,
9 tweets,
2 min read
Read on Twitter
This paper appears to gloss over (or misrepresent) LBNL's annual RPS updates that estimate cost of RPS compliance by analyzing utility-specific spending on RECs and long-term contracts eta-publications.lbl.gov/sites/default/…
This new paper suggests that the LBNL study is based on generation technology costs (LCOE). It's not. The initial report explains the methodology - eta-publications.lbl.gov/sites/default/…. LBNL issues annual updates - emp.lbl.gov/projects/renew…
Totally fair to criticize LBNL's approach, but this new paper appears to ignore it. Instead, this paper uses a regression analysis to estimate costs.
It appears to me that the paper analyzes effects (on generation mix, CO2 emissions) on a per-state basis. But RPS laws don't work that way. Most have regional effects, since they don't generally have state-specific requirements.
And the paper makes it RPS cost claims based on retail electricity prices (following the methodology of the utterly discredited Manhattan Institute paper from several years ago).
Maybe I'm misinterpreting the equations and appendices, but I don't see any effort to control for utility spending on other things (i.e., distribution-level capital expenditures).
The paper says that it controls for "state fixed effects to control for any permanent, unobserved differences across states" in retail prices. Maybe the authors claim this accounts for utility distribution cap-ex?? But focusing on states not utilities is another deficiency.
Paper also assumes a vertically integrated utility model, but I can't tell how that factors into the final results ( section 3). Finally, paper uses 2015 data, which lowballs RPS impact on RE growth - wind and solar were 9% of US gen in 2018; paper says 5% based on 2015 data.
Also, paper outrageously claims that "RPS policies have only increased renewable penetration by a few percentage points." OK. In 2000, wind+solar were 0.1% of US gen (EIA); paper says 5% in 2015...that's a 3,700% increase when you look at MWh.
