SHOCK: Excessive Capacity Market Costs to PJM Customers from Load Forecasting Errors Alone Equal $5 Billion in Single Year, $25 Billion Since 2008/09

The over-forecasting of demand in PJM's capacity market has resulted in an implied consumer cost of $5 billion for the 2013/2014 Delivery Year alone, the PJM Industrial Customer Coalition said in comments to FERC.

As more fully noted below, PJM Industrials said that since the 2008/2009 Delivery Year, overly conservative load forecasts have resulted in over-procurement in the capacity market Base Residual Auctions [BRA] that has cost consumers billions of dollars.

"For the 2013/2014 Delivery Year, consumers have an implied consumer cost due to over-forecasting of approximately $4.94 billion," PJM Industrials said.

"In total, for the 2008/2009 to 2015/2016 Delivery Years, consumers have incurred costs of around $25 billion simply as a result of over-procurement in the BRA due to overly conservative load forecasts," PJM Industrials said.

"This problem needs to be addressed," PJM Industrials said.

"PJM will undoubtedly argue that load forecasting has improved, but citing to improvements does not go far enough to resolve the fundamental problem," PJM Industrials said.

"Consumers simply should not bear billions of dollars in additional costs due to an insufficient and imprecise forecasting model. Procuring additional capacity due to load forecasting errors is inefficient, and the additional capacity that is procured is not worth the cost. RPM has no adequate controls in place to protect customers from over-procurement, and this problem must be rectified," PJM Industrials said.

PJM Industrials explained that the excessive costs arise because capacity clearing prices are consistently overstated in the base residual auctions, largely because of over-procurement prompted by overly conservative load forecasts.

"The conservative load forecasts are exacerbated by fundamental changes to consumption that have yet to be picked up in the forecasting process. For example, consumers have made conscious efforts and investments to be more energy efficient in recent years, and these trends have not been fully incorporated into the load forecasting models. The data from the 2008/2009 Delivery Year through the 2015/2016 Delivery Year highlight this problem. Since the 2008/2009 Delivery Year, load forecasts have been overstated by at least 1.0 percent and reaching as high as 7.6 percent. Since the 2008/2009 Delivery Year, these overly conservative load forecasts have resulted in over-procurement in the BRAs that has cost consumers billions of dollars," PJM Industrials said.