The Roosevelt Project
Electricity market prices across organized wholesale electricity markets in the United States have declined significantly in recent years, prompting several nuclear power stations to consider early retirement before the end of their licensed operation or useful lifespans. This paper explores three possible explanations for observed declines in day-ahead electricity prices received by 19 nuclear generators in the PJM electricity market region: (1) the impact of declining natural gas prices; (2) the growth of wind generation in the American Midwest; and (3) stagnant or declining demand for electricity. I employ time series linear regression with time fixed effects to empirically estimate the effect of each explanatory variable on the average day-ahead locational marginal price (LMP) earned by 19 nuclear generating stations (with 33 individual reactors) located in PJM (encompassing roughly one-third of the U.S. nuclear fleet) as well as weighted average PJM day-ahead market prices. The paper uses daily average observations from January 1, 2008 to December 31, 2016 (n = 3,288). I employ a variety of alternative specifications to further explore geographic heterogeneity in causal effects on different generators across the PJM region and interrogate the impact of using price time series from different natural gas trading hubs. I find that natural gas price declines are the dominant driver of reduced electricity prices at the 19 nuclear power stations over this period. The growth of wind energy has an order of magnitude smaller cumulative effect and is only statistically significant for nuclear generators located in the western portion of the PJM region (in proximity to vast majority of installed wind capacity in the region). Finally, declining demand also has a relatively small but statistically significant effect on prices across all generators.
Keywords: Wind energy, natural gas, nuclear power, merit-order effect, electricity markets, electricity prices, econometrics, time-series regression, fixed effects, energy economics