In his 1971 paper “The Theory of Economic Regulation”, George J. Stigler proposed a framework to consider the motivations of various stakeholders and the influence that they have on regulatory policies. His premise was that the state could use its power to prohibit or compel, through financial restrictions or benefits, to selectively help or hurt industries. Furthermore, the theory of economic regulation should explain the affected parties, form of the regulation and effects on allocation of resources.
Industry regulation and the role of the state
There are two main alternative views on industry regulation. First, there is the view that regulation is put in place to protect and benefit society at large. The second view is that politics are unpredictable and guided by a constantly shifting, diverse mixture of forces. For the purpose of this article, we will assume that the imponderable forces expressed in the second view predominantly influence the decisions of elected officials, such that the public good may or may not be maximized. For example, if an elected representative denies several large industries their intended subsidies or special regulations, these powerful industries will dedicate significant resources to the election of a more favorable successor.
The state has the power to coerce, which is an exclusive resource that is not shared with even the mightiest of its citizens. This coercion can be done either through taxation and redistribution of resources or by ordaining the economic decisions of various stakeholders. From the perspective of industries, when the elasticity of supply is high and there is a threat of new entrants, controls over entry or output tend to be more desirable than direct subsidy of money. This can be done directly through quotas, licenses or other imposed barriers. Alternatively, regulations could be put in place that affect substitutes and complements or fix prices to maintain competitive rates of return.
Distributed energy as a new entrant
Traditionally, the US electricity industry has been dominated by utility companies, which have benefited from a natural monopoly in the transmission and distribution of an electricity commodity. In the late 20th century, the deregulation of power generation introduced competition by allowing independent power producers to bid into the electricity network. Nonetheless, utility companies have continued to benefit from regulated monopoly power and a guaranteed rate of return in transmission and distribution and thus have had a disincentive to allow new entrants such as energy efficiency or distributed energy into the value chain.
Looking at the faster-than-expected penetration of distributed generation, we see a unique case of the new entrants being customers of the incumbent. Because stakeholders with an interest in distributed energy can essentially vote with their pocketbooks on a granular scale, it is extremely difficult to control entry and limit the development of substitutes to the traditional utility monopolies. Furthermore, utility companies are finding themselves competing against their own constituencies. As electricity customers choose to become “prosumers”, or producer-consumers, utilities are faced with a choice between alienating a growing portion of their customer base and allowing their monopoly power to erode gradually.
Coercion of interests through policy
Of course, it should be noted that the emergence of distributed energy–solar photovoltaics in particular–has been supported by policies driven by the demands of environmental interest groups and renewable energy lobbies. More specifically, research grants, renewable portfolio standards, net metering, tax credits and carbon credits have accelerated and sustained the adoption of distributed energy technologies in the United States. Because such policy decisions are coercive, the decision process is fundamentally different from that of a competitive, market-based approach.
In order to influence political decisions, interested voters have to elect representatives by making simultaneous, collective decisions. This decision process involves the entire community, of which only a small minority may be directly concerned about any specific issue. The infrequent and universal nature of participation in the political decision process increases the cost for voters of gathering comprehensive information on the multitude of issues involved. This causes uninformed constituents to affect political decisions with their votes, especially for issues of limited relevance to the majority of voters.
As such, the electoral system lacks the strong incentives of private markets for the acquisition of knowledge to make informed choices. In order to influence the outcome, industry stakeholders can reduce the cost to citizens of acquiring information through education (or uneducation) campaigns meant to rally votes in support or disperse votes in opposition of their interests. The messaging that was used in Arizona media campaigns to debate the merits of solar energy net metering is a clear example of such an attempt.
The dilution of salience and high cost of acquiring adequate information causes votes on issues such as energy policy to trend towards the jurisdiction of state and local governments, where implications of energy policy tend to be salient to a higher proportion of voters. Information asymmetry among voters also makes it difficult to adopt a national long-term energy plan in the United States. These trends are exacerbated by state-level lobbying efforts from interest groups.
Interests of solar energy stakeholders
The following section is a high-level attempt to formulate hypotheses on the primary motivations of solar energy stakeholders.
Solar customers want to benefit from a low cost of solar because they will save money on their energy bill. Given that they vote with their pocketbooks and are subject to education and promotional campaigns from solar service providers, solar customers tend to be well-informed. Their main recourse is to pressure their utility companies, vote in elections and take part in public awareness campaigns.
Non-solar customers bear the burden of subsidies and policies that reduce the cost of solar, such as net metering, through higher taxes and higher electricity prices. These stakeholders are less informed than solar customers and may be unaware of how they are affected. Opponents of solar can attempt to educate, or at times mislead, non-solar customers in order for them to pressure their utility companies, vote in elections and take part in public awareness campaigns against distributed solar generation.
Solar service providers have the most to gain from increased penetration of distributed solar generation. Given that financing, selling and installing solar panel arrays is their core business, a lower cost of solar benefits them through higher sales and more attractive margins. Furthermore, solar service developers take risks in the marketing, financing and installation of distributed solar that result in local learning-by-doing, for which the benefits to future commercial players and consumers should arguably be compensated. Solar service providers can do public education campaigns and invest in industry associations to lobby state and federal government officials in favor of distributed solar.
Balance-of-system manufacturers are very similar to solar service providers, with the exception that their operations tend to be more geographically diversified. As their customers are solar service providers, they benefit from better competitiveness of solar through higher sales. Their risk exposure and R&D efforts to decrease the balance-of-system cost for solar installations results in local learning-by-doing that should arguably be compensated.
Solar panel manufacturers are usually global companies with operations in many countries. As such, they have a smaller stake in individual markets such as California. Nonetheless, a lower cost of distributed solar in one their markets results in higher sales. Solar panel manufacturers invest large amounts of capital to decrease cost per Watt, which results in global learning-by-doing that should arguably be compensated proportionally across geographies. Solar panel manufacturers are a strong force to lobby state, federal and world governments in favor of distributed solar.
Utility shareholders expect stable cash flows at an attractive rate of return on their equity investments. Because the increased penetration of distributed generation threatens to disturb their business model, their default reaction is to limit entry. Although utility companies are regulated natural monopolies that benefit from a guaranteed rate of return, events in Germany have shown that the equity value of utility companies can fall dramatically once distributed generation penetration reaches a certain threshold. Utility shareholders have the means for substantial educational spending and are a strong lobbying force with local, state and federal governments.
Utility managers have a fiduciary duty to their shareholders, employees and customers. We could argue that their main priority is to remain employed and maximize their compensation. As such, their main concern isequity value, which is achieved through stable cash flows, high return-on-invested-capital, low cost of capital and limited risk. In order to achieve this, utility managers need to receive adequate compensation for transmission and distribution services, limit investment, maximize credit ratings and minimize service outages. Utility managers can directly influence state regulators and support industry associations to lobby state and federal governments.
Conventional power providers include the entire fossil fuel value chain from exploration and production to power generation. These stakeholders may lose market power and profits due to a relative decrease inconventional power subsidies and adverse regulatory restrictions. Their main recourse is to lobby through political campaign financing.
Independent system operators are responsible for the efficient and reliable dispatch of electricity into the grid system. The intermittency of distributed solar generation creates difficulties for them in balancing thevariations of power output throughout the day. On the other hand, distributed generation helps reduce congestion and improves reliability at the grid edge. Independent system operators have direct recourse with state regulators.
Financial institutions provide debt and purchase securities in order to make an attractive return for their customers and shareholders. Their main interest lies in the recovery of debt and availability of securities with high risk-adjusted returns. Financial institutions have the ability to vote with their pocketbooks and are a strong lobbying force at all levels of government.
Environmental lobbies act as the watchdogs of global sustainability for all natural ecosystems. Their main concerns with regards to energy are carbon emissions per capita or GDP, carbon dioxide parts per million,climate change in degrees Celsius, air pollution, water pollution, deforestation, waste and biodiversity. Environmental lobbies are a strong force in educating voters and influencing all levels of government.
Labor unions serve the role of protecting workers’ employment and working conditions. Their efforts are mostly centered on domestic manufacturing, job creation, labor force development and collective bargaining. Unions can exert pressure through lobbying, strikes and public campaigns.
Regulators have the task of implementing policies that reduce electricity prices and increase reliability in the electricity system. They follow guidelines set by local, state and federal governments such as renewable portfolio standards and net metering requirements. Regulators also establish the base rate and set the required revenue for utilities.
State lawmakers are elected to office and as such must be attuned to the needs and preferences of their constituents. Their main concerns with energy are reelection, job creation, economic output, air and water pollution, reliable energy and low electricity prices.
Federal lawmakers are elected to office and thus must be sensitive to the needs and preferences of voters. Priorities for them are reelection, energy sovereignty, infrastructure resiliency, economic competitiveness andcarbon emissions per capita or GDP.
Analysis of factors affecting solar penetration
In order to gain an understanding of the influence that stakeholders exert over solar energy policy in the United States, we build a statistical model using state-by-state differences in overall solar energy penetration (% of total capacity) as the dependent variable. For the purpose of this analysis, we compile data on the following independent variables (data table in Annex A):
- Annual insolation
- Average overall electricity retail price (2013)
- Growth of electricity price (2004-2013 CAGR)
- Decoupling of revenues (2012)
- Electricity distribution investment gap (2011-2020)
- Carbon emissions per dollar of GDP (2011)
- Unemployment rate (2013)
- GDP growth (2009-2013 CAGR)
- Number of smog days per year (2010)
- Highest concentration of smog (ppb) (2010)
- Reliance on volatile fossil fuels (2013)
- Use of coal generation (2013)
Conducting a multiple regression analysis on the data for all states yields the following results:
YALL STATES = -0.49 + 0.0006 Exp(X1)+ 0.0044 X2
where
YA = solar penetration (2012 cumulative solar MW / 2010 total MW capacity)
X1 = 2013 average retail price per kWh for all sectors
X2 = year-round average insolation in kWh/m2/day
Observations
- The only variables that produce a statistically significant regression with solar penetration are Insolation (Exp) and Average Retail Price
- We find that 57% of variation in solar penetration among US states can be explained by the objective function with 99% confidence (results in Annex B)
- New Jersey, having by far the highest penetration of solar at low insolation levels, is a clear outlier in almost any combination of variables; we exclude it from the analysis
- Hawaii is excluded as an outlier due to its unique situation being a small island and relying almost exclusively on highly volatile fossil fuels for electricity generation.
Given these results, we conclude that the only measured factor significantly influencing the penetration of solar for all US states collectively are insolation and retail electricity price. These factors are predictable, given that they have a direct impact on the profitability of solar photovoltaics. Interestingly, none of the other variables demonstrate a statistically significant effect on the penetration of solar. This undermines stakeholder effects such as air quality, carbon emissions, unemployment, fuel security or aging infrastructure.
One reason for these results may be that these issues affect various regions differently; in order to test this possibility, we conduct the analysis by dividing the US into three regions based on NERC regional entities. As shown in the map below, we partition the map into West (green), Northeast (blue) and Southeast (red).
Delineations for regional-level analysis
Conducting a multiple regression analysis on the data for each region yields the following results:
West region
YWEST = -0.78 + 0.0008 Exp(X1)+ 0.1019 X2
where
YW = solar penetration (2012 cumulative solar MW / 2010 total MW capacity)
X1 = year-round average insolation in kWh/m2/day
X2 = unemployment as a percentage of total labor (2013)
Observations
- The only variables that produce a statistically significant regression with solar penetration are Insolation (Exp) and Unemployment
- We find that 88.7% of variation in solar penetration among states in the West region can be explained by the objective function with 99% confidence
- Hawaii is excluded as an outlier due to its unique situation being a small island and relying almost exclusively on highly volatile fossil fuels for electricity generation.
Northeast region
YNORTHEAST = -0.46 + 0.0053 X1
where
YN = solar penetration (2012 cumulative solar MW / 2010 total MW capacity)
X1 = 2013 average retail price per kWh for all sectors
Observations
- The only variable that produces a statistically significant regression with solar penetration is Average Retail Price
- We find that 23.7% of variation in solar penetration among states in the Northeast region can be explained by the objective function with 90% confidence
- The results are highly heteroscedastic (see Annex D), suggesting that drawing conclusions from the least squares regression may be unreliable.
Southeast region
YSOUTHEAST = -0.58 + 0.0066 X1
where
YS = solar penetration (2012 cumulative solar MW / 2010 total MW capacity)
X1 = 2013 average retail price per kWh for all sectors
Observations
- The only variable that produces a statistically significant regression with solar penetration is Average Retail Price
- We find that 49.7% of variation in solar penetration among states in the Southeast region can be explained by the objective function with 99% confidence.
Based on these results, we conclude that a regional level analysis of the factors affecting solar penetration yields slightly different results than when including all states simultaneously, particularly in the West region.
West region: Given that the West region spans from North to South, it seems intuitive that insolation is an important factor for solar penetration. Surprisingly, the retail price of electricity has no significant effect on penetration. The unemployment rate, however, has an important effect. This is the only case we found where a variable other than retail price or insulation can be included in the objective formula. Together, insolation and unemployment account for almost 90% of the variation between states in the West region with 99% confidence.
Northeast region: Due to the retail price accounting for less than 25% of the variation and the presence heteroscedasticity in the regression, we conclude that the available data does not adequately explain the factors influencing solar penetration in the Northeast region.
Southeast region: Average Retail Price alone explains roughly half of the variation between states in the Southeast region with 99% confidence.
Conclusions
Based on our analysis, differences in solar penetration between states can overwhelmingly be explained by factors affecting the economic competitiveness of solar, such as insolation or retail electricity prices. Our hypothesis was that the motivations of multiple stakeholders shape the policies that affect the adoption of solar energy; given the variables we have examined, we find that the data is inconclusive. It should be noted, however, that when regressing penetration of solar in all US states with our data set, we found that 57% of the variation could be explained by a combination of insolation and average electricity retail prices. Accordingly, 43% of the variation may be attributable to other factors such as stakeholder motivations not included in this analysis.
Comparing states within regional groups shows that results can vary significantly between different parts of the country. In the West region, we found that a combination of the insolation levels and unemployment rates predicts solar penetration with high accuracy. The statistical significance of unemployment in this case is notable, as this is the only instance where a variable other than insolation or retail electricity prices fit the regression. One explanation may be that states such as California and Nevada, which have the highest unemployment rates in the region, see strong benefits in leveraging their geographic advantage to become solar industry leaders.
In order to improve the analysis, we would consider changing the dependent variable; the amount of subsidization ($/kWh) in each state, for example, would more accurately measure the influence of stakeholders on energy policy. Alternatively, regressing stakeholder motivations with barriers to entry such as permitting or interconnection delays may yield some interesting insights. A hypothesis may be that different stakeholders affect the penetration of solar in different ways. If this were true, stakeholder motivations may affect outcomes such as subsidies and barriers to entry, which would in turn impact the penetration of solar generation.
As distributed solar generation continues to gain traction in the United States, the debate on the future of energy will become salient to an increasingly large constituency of voters. The fate of this critical part of our national infrastructure will undoubtedly be shaped by the motivations of concerned stakeholders. In this article, we attempted to glean actionable insights on the impact of these stakeholders. Although the factors that translate their motivations into coercive energy policies remain a mystery, we did learn one lesson: pocketbook votes are worth much more than polling booths.
ANNEX A
Sources (from left to right):
U.S. Solar Market Trends 2012, IREC, 2013 / EIA-860
U.S. Solar Resource Maps, NREL, 1998 to 2009
Electric Power Monthly, EIA, August 2013
Form EIA-826 Detailed Data, EIA, 2001 to 2010
U.S. Climate Policy Maps, Center for Climate and Energy Solutions, 2013
The Economic Impact of Current Investment Trends in Energy Infrastructure, ASCE, 2011
State Energy Consumption, Price, and Expenditure Estimates, EIA, 2011
U.S. Bureau of Labor Statistics, 2013
U.S. Bureau of Economic Analysis, 2009-2013
Danger in the Air: Unhealthy Air Days in 2010 and 2011, Environment America, 2011
Ibid.
Electric Power Monthly, EIA, July 2013
Ibid.
ANNEX B
Results of regression (all US states)
*Outliers: Hawaii, New Jersey
ANNEX C
Results of regression (West)
*Outlier: Hawaii
ANNEX D
Results of regression (Northeast)
ANNEX E
Results of regression (Southeast)
Leave a Reply