Anyone get a sinking feeling when applying market data averages to future cleanenergy investments? Fair warning, this is a long – but surely riveting – post about statistics in climate finance.
Despite being a common occurrence in the industry there are a lot of pitfalls in applying market data to proformas and forecasts. And the downside is steep - a 10% variance in an estimate could eliminate the entirety of contingency funds.
By way of example, Vivint Solar - over the period of 2019-2021 - reported+70% of their installed systems in Massachusetts had cost exaclty $3.46 per watt to install. Likely the data was submitted as a proxy averaging the portfolio of projects rather than sharing real costs. if the case, the submission would result in thousands of proxy figures alongside real costs.
Why should we care?
(i) Variance in Market Data: Every clean energy project is unique with variables like geographical location, design complexity, technology choices, and more. It is essential that we recognize the high degree of variability within the market data and inherent risk in applying 'average' numbers to our projections. The Vivint data disguises this variance that can otherwise be seen from other installers in the graphic.
(ii) The Deceptive Nature of Averages: Averages can often camouflage the actual distribution and variance from prior experience. This is especially problematic when planning solar energy projects as large variances could mean the difference between project feasibility and a financial loss. Without scrutinizing the data more deeply removes the opportunity to identify and mitigate risk. Very few projects result in the average, as the distribution shows.
(iii) Extrapolation and Future Uncertainty: Technology advancements, policy changes, supply chain disruptions, and many other factors can affect future costs in unpredictable ways. A prerequisite of accurate projects is the constant refresh of data to maintain its relevance and accuracy. The data shown ends in 2021, and may not account for recent changes in market forces such as inflation.
Comments