Photo: Mel Watkins

Here is the latest installment in our continuing series of commentaries celebrating the 50th anniversary of the publication of Mel Watkins’ classic article, “A Staple Theory of Economic Growth.” This commentary is from Mel’s long-time collaborator Thomas Gunton, Director of the Resource and Environmental planning Program at Simon Fraser University. Gunton’s submission, supplemented by an extensive bibliography, applies staples analysis to the current boom in resource-oriented petroleum developments in Canada, and finds the staple theory to be as powerful and relevant as ever in explaining Canada’s economic and political trajectory.

Fifty years ago, Mel Watkins provided a brilliant articulation of staple theory that was instrumental in stimulating a productive staple tradition in both Canadian and international literature over the last half century. This is indeed fortunate because there is little doubt that the staple tradition remains as relevant to Canada today as it was 50 years ago.

Canada is currently embarking on one of the largest staple-based economic expansions in its history: including the rapid expansion of bitumen production in Alberta, and the planned development of liquefied natural gas (LNG) exports from B.C. The magnitude of this expansion is dramatic. Oil production is forecast to more than double from 2010 to 2020 (Millington et al. 2012), and B.C. is contemplating at least five new LNG facilities involving $98 billion in investment and a three-fold increase in natural gas production by 2020 (Ernst and Young 2013).

The environmental implications are also staggering. Greenhouse gas emissions resulting from oil expansion will increase three-fold, from 45 Mt/year in to 155 Mt/year by 2030 (Millington et al. 2012), while other emissions will negatively impact air, land, and water (Joseph et al. 2013; Grant et al. 2013). Data from Environment Canada show that without oil sands expansion, Canadian greenhouse gas emissions would actually decline by 5.1 per cent between 2005 and 2020 (Environment Canada 2013, p. 21). Cost-benefit analysis of oil sands development shows that when environmental impacts are included, oil sands development is a net cost to society (Joseph et al. 2013).

Given the magnitude of the expansion and its impacts, it is important that Canadians have a comprehensive understanding of this development. Unfortunately, most of the analysis to date lacks a political economy perspective. Publications from the oil and gas sector focus on the alleged economic benefits and the need to remove “barriers” to expansion (CAPP 2013; Honarvar et al. 2011). Governments are engaged in advertising campaigns extolling the virtues of oil and gas development, while much of the criticism of the expansion focuses on specific themes such as greenhouse gas emissions. While these critiques provide important insights, they do not provide a comprehensive understanding of the overall dynamics of staple development.

Thanks to the efforts of Watkins and the Canadian staple tradition he helped to spark, we have a rich analytical framework for understanding the current oil and gas boom. While a comprehensive analysis is beyond the parameters of this short overview, it is useful to provide a brief outline of how staple theory would be applied.

The starting place is Watkins’ 1963 article. Watkins explains how external demand stimulates growth through the extraction of the staple, and through the series of spread effects that he divides into the well-known categories of forward, backward, and final demand linkages. Later Watkins and others (Kierans 1973; Gunton and Richards 1987) added what Hirschman (1981) terms the fiscal linkage, defined as the “rent” generated by the staple. The production function defines the potential linkages of the staple. Growth can occur over time; the economy may either diversify as it surpasses critical thresholds for new activities (the optimistic paradigm) or stagnate if and when the staple sector declines due to changes in demand and/or exhaustion of the staple (the pessimistic dependency paradigm).

Watkins identifies key factors that influence the course of this development. Some such as external demand, quality of the staple, and the production function are exogenous variables largely beyond the control of the staple producing region. Other factors such as the ownership of staple industries and the distribution of rent are endogenous — in the sense that they could potentially be managed by the staple region, albeit with great difficulty.

One endogenous factor cited by Watkins is foreign ownership, which can inhibit growth by leaking income from the staple region to compensate foreign owners and by relying on foreign-located forward and backward linkages instead of developing them within the staple region. Another factor is an “export mentality” that leads the staple region to concentrate on the staple to the exclusion of alternative investment opportunities. This mentality engenders excessive optimism that leads to overinvestment in the capital-intensive staple, leaving a legacy of rising debt that is increasingly difficult to service in the face of declining commodity prices and rising production costs (as the most productive staple resources are exhausted). The result is a severe and protracted downturn and a period of painful readjustment.

Watkins also highlights the potential negative impact of staple development on the subsistence or domestic economy. According to the staple model, governments can become subservient to the staple sector, focusing on stimulating short-term growth by subsidizing development.

What insights can staple theory provide about the current oil and gas boom? Conventional predictions of the economic effects of oil and gas developments are used uncritically to demonstrate the contribution of the sector to the Canadian economy. CERI (Millington et al. 2012), for example, in a recent study estimates that new oil sands projects will generate $2.1 trillion in economic activity and almost one million person years of employment between 2012 and 2035. The assumption is that such growth is beneficial.

Staple theory takes a more analytical approach and asks whether this expansion indeed contributes to the long run sustainability of the economy. How does forecast growth compare to potential growth both in terms of magnitude and structure? Are the potential spread effects fully exploited — and if not, why not? Although the answers to these questions require detailed analysis to estimate actual relative to potential effects, available evidence suggests that the linkages in the oil and gas sector are not well developed and consequently the potential contribution of oil and gas development to Canada is not being maximized.

The industry’s own forecasts show that almost all of the new oil production will be exported in an unprocessed state to be refined in foreign markets (CAPP 2012). Raw bitumen will not even be upgraded into crude in Alberta, despite the fact that upgrading reduces shipping costs by avoiding the need for diluent, which is necessary to ship bitumen in pipelines. Crude is also more marketable than bitumen because of the larger number of refineries that can process crude (Enbridge 2010). Many of the backward linkages are also not developed in Canada. Ironically, the Alberta government’s own website provides a “US Economic Impact Oil Calculator” that documents the industry’s reliance on U.S. suppliers, and concludes that one-third of all the jobs generated by Canadian oil sands expansion will be created in the U.S. As the website states:

Refineries across the U.S. are hiring workers to build new units that can process oil sands output. Most of the giant trucks used to produce the oil sands are manufactured in Illinois. Much of the software to run the complex production systems comes from California’s Silicon Valley. The oil sands industry is demanding leading-edge water treatment systems, environmental technology, hydrocarbon processing equipment, and countless other goods and services, much of it imported from across the U.S. (Alberta 2013).

There are many reasons why these linkages are not developed in Canada. Some may be based on economic fundamentals such as the availability of low-cost surplus refining capacity in the U.S. (relative to higher-cost new capacity in Canada). Some may be due to the institutional bias of foreign firms to supply existing refineries that they own, instead of building competing facilities within the staple-producing region. As discussed above, the rationale for upgrading bitumen into crude to reduce transport costs (by up to 60 per cent) and increase the marketability oil is persuasive. The role of these and other factors requires detailed analysis. However, unlike conventional analysis that simply accepts the outcomes as a given, staple analysis asks what the actual and potential spread effects are, what factors determine the spread effects, and what can be done to strengthen linkages within the staple region.

Staple analysis also focuses on the fiscal linkage. How much rent is generated by the staple, and where does the rent go? These are important questions that address both the contribution of the staple to development — which is a function of the proportion of rent reinvested back into the staple region relative to the proportion leaked — as well as the equity questions of who gains and who loses from staple development. Maximizing the contribution of the staple requires collecting rent and distributing it equitably to the owners of the staple; in the case of oil and gas, this is the public in the producing regions.

Again, available evidence suggests that a large proportion of rent is foregone by the public owner and retained by private sector oil and gas companies as a surplus return above normal returns to private capital. Plourde (2010) estimates that the private sector retains between 38 per cent and 65 per cent of the rent even under the new more aggressive Alberta royalty regime. A more recent study by the oil sector using a different methodology concludes that the private sector retains 65 per cent of the revenue from an oil price increase (Egglington et al. 2012), while a cost-benefit analysis of a major project estimates rent retained by the private sector is in the range of 35 per cent (Joseph et al. 2013). Given that 47 per cent of the profits to the oil and gas sector accrue to foreign owned companies (StatsCan 2012), most of the rent retained by the private sector is ultimately leaked to foreign owners.

Another economic impact addressed by staple theory is the effect of the staple expansion on other sectors. While conventional analysis considers only the positive linkages, staple theory looks at both the positive and the negative. Expansion of the oil and gas sector can negatively impact other sectors by using labour and capital that would otherwise be available to support growth elsewhere in the economy. Expansion of oil and gas can also undermine the competiveness of other sectors by raising the exchange rate: the so-called Dutch disease. Estimating the negative impacts of oil and gas expansion on other sectors is complex and subject to many uncertainties. Nonetheless there is little doubt that a significant proportion of the labour and capital employed in the oil and gas expansion would have been employed elsewhere in the Canadian economy in the absence of oil and gas expansion. Labour and capital have in economic terms an “opportunity cost.” There is also ample evidence that the expansion of the oil and gas sector has raised the value of the Canadian dollar and that this has resulted in a reduction in the Canadian manufacturing sector (Bimenyimana and Vallee, 2011; Campbell, 2011; Clarke et al. 2013; OECD 2012; Woynillowicz and Lemphers, 2012).

As Watkins illustrated in his work for the Dene (Watkins 1977), staple theory is also useful in understanding the impact of staple development on First Nations. If First Nations are integral to the extraction of the staple (such as in the fur trade), they may receive some benefit from staple economy. If they are obstacles to staple development, First Nations are more likely to be excluded and potentially harmed to the degree that their traditional economic activities are negatively impacted by staple extraction.

The oil and gas sector is an increasingly complex environment for First Nations. First Nations traditional subsistence activities are clearly jeopardized by environmental impacts of oil and gas development. Contamination of the environment by release of toxic chemicals and oil spills threaten First Nations water and food supply, and current compensation and regulatory regimes are inadequate to protect First Nations interests (Gunton and Broadbent 2012). Aboriginal rights and title affirmation by Canadian courts does provide some protection for some First Nations. While First Nations possessing rights and titles do not have clear veto power over resource development, the oil and gas sector is required to engage in meaningful consultation and seek accommodation of First Nations interests in resource development plans. This has led to the emergence of Impact Benefit Agreements that provide direct benefits to some First Nations (Coates and Crowley 2013).

Another key component of staple analysis is the role of government. Governments in staple theory are not independent actors rationally choosing policies based on the “public interest” as assumed in government legislation governing oil and gas expansion such as the National Energy Board Act and the Alberta Responsible Energy Development Act. Instead, as the dominant sector driving economic activity, staple industries are able to successfully pressure governments to implement policies that serve the interests of the staple sector. Governments become promoters of staple expansion by setting weak regulatory regimes conducive to development, providing infrastructure, expediting regulatory review processes, providing marketing support, and “managing” opposition to staple expansion.

Again the evidence on the oil and gas sector lends support to the staple theory analysis of government. The federal government, for example, recently amended Canadian regulatory legislation to expedite the approval process for pipelines and restrict the role of potential opponents in the hearing processes (Westcoast Environmental Law and Ecojustice 2012). Measures to address greenhouse gas reductions have been delayed or, in the case of the Kyoto Accord, abandoned. Caps are set on liability for oil spill damages to protect the oil and gas sector (Gunton and Broadbent 2012), while promotional programs extolling the virtues of the oil and gas sector are being implemented and aggressive lobbying efforts to secure access to foreign markets, such as the Keystone XL lobby in the U.S., are being implemented. The government attacks opponents of expansion as foreign radicals (Oliver 2012). And as discussed above, governments have done a poor job in collecting rents for the public owners and have subsidized the expansion of the oil and gas sector by over $2.8 billion per year (Sawyer and Stiebert 2010).

The staple theory’s contribution to understanding the development cycle is another key contribution. Preceding this year’s Nobel laureate Robert Shiller’s theory of “irrational exuberance” by 50 years, Watkins warns in his 1963 article of the “excessive optimism” in the staple sector that causes booms to proceed beyond their proper limits, generating excess capacity that leads to downturns and financial collapse.

Previous resource booms such as the post-1979 petroleum boom exhibit this excessive optimism dynamic (Gunton 2003; 2004). The current boom displays similar features. As discussed above, the Western Canadian oil sector is undertaking an unprecedented expansion and the federal and provincial governments are urging even more rapid expansion to avoid “missed opportunities” in export markets. The result is that prices for Canadian oil exports are already being depressed by the too rapid expansion of the sector — the so-called bitumen bubble — as regional markets are unable to absorb the increased output (Egglington et al. 2012; Moore et al. 2011). A further doubling of production will do little to stabilize export prices. The large number of LNG projects being considered for B.C. exceeds forecast demand and is based on a large differential between North American and Asian prices that may not last, leaving B.C. with uneconomic projects (Gunton and Broadbent 2013). The Canadian industry is a relatively high-cost producer and therefore is highly vulnerable to the inevitable downturns in markets. Staple theory provides a clear warning of the dangers of such excessive optimism and the need for a more prudent approach of slower sustained growth.

Ownership is another key theme in staple theory. As Watkins illustrates, ownership affects staple development by impacting the location of spread effects and impacting the development of domestic entrepreneurship. Again, very little analysis of the Canadian oil and gas sector addresses the role of foreign ownership. Almost one-half of oil and gas companies in Canada are foreign-owned (Statistics Canada 2012), and incremental oil production is being used to supply refineries outside Canada. Some oil is being traded within existing companies according to intra-company transfer prices, which makes the collection of royalties challenging. This is particularly true for lower-value products such as bitumen for which there is not a well-developed market price that can be used to determine royalty payments (ARRP 2007). In fact many existing Alberta oil producers directly benefit from the depression of Canadian export prices by realizing the benefit of higher refinery profits (due to lower crude supply costs) while paying reduced royalty payments to Alberta. And what incentive do state-owned companies such as China’s CNOOC have in maximizing the export price of Canadian crude or developing refining capacity in Canada? Even the federal government seems to be belatedly aware of the implication of foreign state-owned enterprises, as represented by its stricter takeover policy implemented after approval of the takeovers of Nexen and Progress Energy in 2012 (Harper 2012).

What policies should be implemented to maximize the contribution of staples to Canada? Again staple theory provides clear and persuasive guidance. According to staple theory, a successful strategy should include the following key components:

– collect rent and ensure that it is reinvested back into the Canadian economy

– develop linkages where economically feasible to maximize the spread effects. For example, there is a strong case for at least upgrading bitumen prior to export to reduce transportation costs and increase the marketability and hence net return

– maximize long-run rent by slowing the rate of expansion to a sustainable pace to help moderate construction costs, prevent downward pressure on export prices, and reduce the negative impacts on other sectors of the economy

– mitigate environmental costs by reducing greenhouse gas and other emissions, making Canadian oil and gas the cleanest fossil fuel in the world

– fully engage impacted First Nations in development planning and preparation of impact benefit agreements that ensure and equitable distribution of benefits

– invest in training Canadian workers for oil and gas employment

– phase out all government subsidies to the Canadian oil and gas sector

– restrict foreign takeovers and strengthen domestic ownership of the oil and gas sector.

As staple theory cogently illustrates, developing staples is a challenging endeavour filled with both opportunities and pitfalls. As Canada embarks on its current expansion of oil and gas, Canadians need to assess the degree to which this expansion serves Canada’s interest. And thanks to Mel Watkins, we have a powerful analytical framework for undertaking such an assessment. We can only hope that more researchers follow the lead of Clarke et al. (2013) and use staple theory to analyze the current boom in Canadian resource developments. This would improve our understanding of the current staple boom, and hopefully encourage a more prudent approach to developing Canada’s oil and gas sector.

A list of references cited in this article is available here.