A public model on pricing solar PV versus Coal

Blowing Hard or Shining Bright? Making Renewable Power Sustainable in India

Content from the Brookings Institution India Center is now archived. After seven years of an impactful partnership, as of September 11, 2020, Brookings India is now the Centre for Social and Economic Progress, an independent public policy institution based in India.

This model accompanies a chapter on the “Economics of Renewable Power: It Depends” in the book Blowing Hard or Shining Bright? Making Renewable Power Sustainable in India, edited by Brookings India, also available as a standalone paper. The chapter aims to discuss the economics of Solar PV pricing with coal-based power pricing in 2014 in India, through a transparent multi-variable model with parametric inputs spanning a range of plausible and applicable values.

Lies, Damn Lies, and Assumptions

With apologies to Mark Twain, who famously stated “Lies, damn lies, and statistics,” this note exemplifies the strong need for transparency in assumptions as an important facet of models and calculations, on which decisions could be made.  Lack of transparency (or error-checking) led to the simple mistake of mixing the metric system and imperial system, leading to the crash of NASA’s $125 million Mars Orbiter.  Closer to home, we have a heuristics for whether to build a pit-head coal plant or transport coal, with the cross-over cited as 800 km, even though that number is based on decades old calculations. Even more subtle than outdated calculations is the use of assumptions that favour a particular outcome.  Sometimes, choices of assumptions can be scrutinized, understood, and dismissed easily (e.g., the belief that 100% of available “waste” land becomes available for renewable power), but many other times, the assumptions themselves have uncertainty, and sometimes values become outdated or inappropriate for only a particular case. It is in this spirit that a model for the comparative analysis (illustrative only!) for solar photovoltaic power versus coal-based power in 2014 is made available publicly, so that those who wish can change any assumption to see the implications.  While most variables can be changed, this model framework itself is based on the use of levelized cost of energy (LCOE), which is a useful tool but has its limitations.  The intent isn’t to claim any particular answer or value, but to show plausible ranges and highlight what input factors really matter.  Ultimately, such models and their outputs need deliberation and perhaps further analysis.

Don’t like the answers? Change the input values

Click here to download the model, and anyone is free to experiment with it, under attribution creative commons licensing. The main model is in the .ana or Analytica framework, which is a decision and modeling analysis tool from Lumina Systems.  While the advanced version is for sale, a free version of the software is available for anyone to download, which is only limited by the number of nodes it can handle.[1]  This model thus can be run by anyone, anywhere, and the software is intuitive and easy to run. Why use a non-typical software instead of widely understood and available ones, such as Excel? The reason is that Excel cannot easily handle the scale of parametric analysis as used in this relatively simple model (roughly 50 nodes, of which 15 can be varied by the end-user).  The model is kept relatively simple, but Analytica is a tool for not just parametric analysis, but also stochastic modeling (Monte Carlo simulations), all using an intuitive Influence Diagram Graphical User Interface.  Thus, for example, if one wanted, instead of modeling imported coal prices for the base year as either one out of the choices of {70, 80, 90, 100} $/ton, one could assign a probability distribution based on a mean, variance, and distribution shape if one wished. Most importantly, this tiny file (~29 kB) simultaneously outputs roughly 15 million possibilities in a parametric space (15 variables, averaging 3 choices each, perhaps more, outputting at least 315).  While the newest versions of Excel could handle this size, managing, manipulating, and comprehending it is slow, even with PivotTables.[2]  However, to ensure those unable to use Analytica can try the model, it is also available in an Excel format for download (where variables can be changed one by one only, instead of simultaneously). The results of the model and assumption details are available in the Chapter Renewables and Economics (and standlone paper). The chapter is part of Brookings India’s edited book “Blowing Hard or Shining Bright? Making Renewable Power Sustainable in India”. To see the full book please click here. 

Click here to download – Excel Version

Click here to download – Analytica

[2] On a fast machine, Analytica can calculate the results in approximately 30 seconds.