Benchmark base year vs. corp disclosure base year

[MichaelTiemannOSC]

In the world of corporate disclosures, targets are often stated in terms of base years (which may date back as far as 1990), start years (when the target was published) and end year (when the target reduction ambition is hopefully achieved).

In the world of benchmarks, there is a start year (typically 2019) where we begin with status quo, and an end year (typically 2050) which is the projection limit of the benchmark.

The original SBTi code has a concept that a company has some sort of "base year production" for its S1+S2 emissions (ghg_s1s2) and its S3 emissions (ghg_s3). As I have understood the code, these base year values tie to the production controls, which tie to the benchmark, not the target data definition of base year. It's somewhat messy.

I am trying to resolve a problem where a company has no production data for 2019, but does have production data for 2020 onwards. I see three options:

1. Drop the company due to lack of data
2. back-cast data to 2019 (either by copying 2020 data or trying to follow its trajectory backward)
3. fix the code to deal with ragged left-side data (not easy due to initial assumptions in original SBTi code)

If option 3 is the preferred option, I can write up some documentation explaining that this doesn't work yet but an ambition for a future release. Otherwise I can document how the other options deal with the data default. Presently the tool does log an error when such production data is missing (so users know it's missing), but it tries to carry on and I need to decide whether to carry on without or with an approximation.