Future Renewable Energy Demand: It's the Economy (and DSM), Stupid!

Ryan Hubbell's picture

Not long ago, continued growth of electricity consumption in the U.S. was as certain as population growth. With historical data trending upwards linearly, it would seem a fool's errand to arrive at any other conclusion. Yet I seem to recall a couple other major industries relying on extrapolation recently (real estate valuation ring a bell?), and we're nearly bankrupted as a result.

Electricity generation by fuel type (natural gas, renewable, coal, nuclear, and other) in trillion kilowatthours per year from 1990 – 2035. Historical and projected growth for all fuel types except coal and nuclear which decline slightly, from 2010 – 2035.

Figure 1. Electricity generation by fuel (trillion kWh per year);
Source: EIA 2012 AEO preview

U.S. per capita electricity consumption in kilowatts per year and U.S. per capita GDP in constant 2000 USD from 1960 to 2009. Both trend up until 2008. Lines diverge until 1975 then both trend upwards equally spaced, following the same relative pattern.

Figure 2. Electricity Consumption and GDP per capita;
Source: Environmental Defense Fund

Now, I'm certainly not suggesting the energy industry is similar to, or will suffer the same fate as, the finance or insurance industries. But manufacturers and developers banking heavily on continued near-term expansion of electricity generation may be in for a reality check. This is especially true for renewable energy generation. According to the U.S. Energy Information Administration's 2012 Annual Energy Outlook—Early Release, electricity generation in the U.S. is projected to remain at or below 2008 levels until 2016 (Figure 1). The reasons for this decline are not necessarily numerous or complex. As the saying goes, "it's the economy, stupid."

Although many factors influence aggregate electricity consumption, nearly all can be linked to the economy. Figure 2 illustrates the relative correlation between per capita GDP and per capita electricity demand. This is not surprising when you consider that roughly 75% of total electricity consumption comes from the commercial and industrial (C&I) sectors. Economic activity also influences local population and income levels, both of which drive housing levels, and in turn, electricity consumption. Thus, as economic activity decreases, so does electricity consumption.

Another driver of declining electricity consumption is utility demand side management (DSM) programs. Xcel, for example, attributes its 700 MW downward revision of planned firm obligation load through 2018 to its DSM program in Colorado. (The other major factor cited was an anticipation of lower economic activity.)

A reduction in load results in reduced energy sales, in part because electricity is currently not stored long-term on a utility scale. Given that renewable portfolio standards (RPS) are typically based on a percentage of total electricity sales, a decline in load also means a decline in wholesale demand for renewable energy to go towards meeting the RPS. With a reduced need for additional capacity, utilities have less incentive to enter into future power purchase agreements, which—as any developer knows—are key to financing utility-scale projects. And that's not all.

If the economy remains supressed, end users will likely continue to seek ways to cut electricity costs. This may cause utilities to increase rates because major infrastructure investments are typically recovered over a 20-year period based on certain forecasted sales levels. In other words, if sales go down, prices could go up because the fixed infrastructure costs must then be allocated over a fewer number of megawatt hours in sales. This is especially true in regulated markets where utilities are permitted to achieve specific cost recovery rates.

So a number of questions remain that will have a significant impact on future electricity demand. Will DSM and a slow economy create a reinforcing loop between lower consumption and increasing electricity rates? Or, will lower natural gas prices reduce operating costs enough to cover lower electricity sales? Will the economy pick up and stimulate electricity demand? Or will distributed generation decouple GDP and electricity consumption (or load) per capita?

Given all the uncertainty, one thing does seem probable: history will not be as good an indicator of future load as it was in the past.