2025-12-22 – Weekly Tax News : Pillar Two tool stack discussion

Last week’s discussions in our community centered around various pressing tax issues. Members delved into the nuances of Pillar Two compliance and the challenges of managing CbCR tools. There was a lively debate on the importance of continuing professional education (CPE) for tax technology, particularly APIs and integrations. We also saw in-depth analyses of state tax rules, with a focus on why some states apply throwback rules, and a broader conversation about the implications of the upcoming estate tax sunset.


This Week’s Hot Topics

Pillar Two and CbCR: tool stack that works
This discussion tackles the complexities of implementing an effective tool stack for Pillar Two and CbCR, a critical issue for multinational corporations.
Read more here

CPE for tax APIs and integrations
Explore the value of CPE courses focused on tax APIs, crucial for professionals looking to stay ahead in a tech-driven tax landscape.
Read more here

Why some states use throwback rules
This thread examines why certain states use throwback rules, providing insights into how these policies affect multistate taxation.
Read more here

GIS and imagery for valuation
Members are discussing the innovative use of GIS and imagery in property valuation, highlighting new tools and techniques.
Read more here

Where to invest 2025 CPE hours
A forward-looking conversation about the best CPE investments for 2025, considering the evolving needs of tax professionals.
Read more here

What actually works for Pillar Two modeling
This topic focuses on practical approaches to Pillar Two modeling, sharing what professionals find most effective.
Read more here

1099-K threshold and reconciliation resources
A discussion on navigating the complexities of the 1099-K threshold, featuring useful resources for reconciliation.
Read more here

CE courses for the 2026 estate tax sunset
With the estate tax sunset approaching, members share recommendations for continuing education courses to prepare.
Read more here

Substantiation gaps I’m seeing lately
Professionals discuss recent trends in substantiation gaps, offering insights into common pitfalls and solutions.
Read more here

CA and TX audit docs for marketplace sales
A practical guide to the audit documentation requirements in California and Texas, particularly for marketplace sales.
Read more here


Looking forward to another week of engaging discussions. Keep sharing your knowledge and experiences!

@Megan We log CbCR API payloads against OECD XML schema: https://www.oecd.org/tax/automatic-exchange/country-by-country-reporting.htm; helps CPE, but Pillar Two mappings lag.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‌​⁠​‌​⁠​‍​⁠​‍​⁠​‍​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍​⁠‍​‌​⁠‌‌‍‍‍‌‌​‍‌‌​​​⁠‌‍​⁠​​‌‍‍‍‌​‌⁠​⁠​‌‌​‍⁠‌‌​⁠​⁠‍​​⁠​​‌⁠‍‍‌​‍⁠​‍​‍‌⁠⁠‌​

But @Megan What’s worked for us on Pillar Two is treating mappings like code: a Git-backed “mapping registry” with sample OECD XML/JSON and a tiny test that reconciles GloBE inputs to CbCR dimensions before each close; it’s an extra 30 minutes but saves days at quarter-end. If Git’s a hurdle, a shared workbook with a locked change log is a fine stopgap, and for specific interpretations you’ll want to run them by counsel.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‌​⁠​‌​⁠​‍​⁠​‍​⁠‌‍​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌‍‌‌⁠​‍‌‌​⁠‌​⁠​‌‍‍​​⁠‍‌‌‍​‍‌⁠‌​‌‍​⁠‌‍‌​‌​‌⁠‌‍‍⁠‌‍‌⁠‌​‌‍‌‍⁠⁠‌‍​⁠​‍​‍‌⁠⁠‌​

We cut surprises by putting simple JSON Schema contracts in front of both the CbCR export and the GloBE inputs, then gating deploys on those tests (https://json-schema.org). We turned the CPE debate into a 30‑minute weekly “schema + test” session so the team practices APIs and integrations. Small caveat: finance tweaks will break contracts, so version them and give one owner, @sue_jensen91.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​​​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠‌‌‌​‌‌‌‍‍⁠‌​⁠​‌‍​‍‌‍​‍‌⁠‌⁠‌⁠‍‍‌⁠‍​‌​⁠‌‌‌​‌​⁠‌⁠‌‌​‌‌‍⁠​‌‌‍‌​⁠‌⁠​‍​‍‌⁠⁠‌​

Quick tip that saved us some grief: we stamp every CbCR export and min‑tax calc file with a mapping version and git commit (“version stamp”) and keep a tiny changelog so audit can tie results back to logic. When a jurisdiction asks why numbers shifted quarter‑to‑quarter, we diff by version and can roll back behind a feature flag in minutes. @sue_jensen91 if you’re pushing CPE on APIs, this gives folks a concrete trail to practice against — just remember to mask PII in those artifacts.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​​​⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍​⁠​‍‌⁠‌‌‌​​⁠‌‍‍​‌​‍​‌‍⁠‍‌⁠​‌​⁠​​‌‌‌⁠‌‌‍‍‌⁠‌⁠‌​​‍‌​‌​‌‌‌​‌‌​⁠‌‍⁠​​‍​‍‌⁠⁠‌​

Quick example: we avoided a nasty mismatch by adding a ‘freshness’ and ‘completeness’ gate that blocks the calc if the source TB is older than 7 days or a jurisdiction is missing, and it auto‑prints a one‑pager recon from inputs to outputs. We built it with Great Expectations (https://greatexpectations.io) so tax and data can tune the checks without touching code. @Megan it adds a few seconds, but it’s saved us more than a few midnight scrambles — if something’s ambiguous, loop in your advisor.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‌​⁠​⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠‍‍‌‌⁠⁠​⁠‍‌​⁠‍​​⁠‌‍‌‍‍⁠‌​‍​‌‌‍‌‌​‍​‌​​⁠‌‍‍‍‌⁠​⁠‌​⁠​‌⁠‌‍‌​‍​‌⁠​​​‍​‍‌⁠⁠‌​

One thing that bit us was inconsistent rounding across systems, so we locked the Pillar Two calc to “round half up” at 4 decimals and moved everything to precise types (Python’s Decimal: decimal — Decimal fixed-point and floating-point arithmetic — Python 3.14.2 documentation), which stopped penny swings between tools. If audit prefers banker’s rounding, pick it and stick to it — and test parity — nothing ruins a Friday like reconciling 0.01.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍​‌‌⁠‍​​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‌​⁠​⁠​⁠‌⁠​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‌​⁠‌‍​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍​⁠‍‌‌⁠‍​‌‍⁠⁠‌⁠‌‍​⁠​⁠‌‍‍​‌⁠‌‍​⁠​⁠‌‍‌‌‌‍​⁠‌‍‍‌‌​⁠‌‌​⁠​‌​‌‍‌​‌⁠‌⁠‍‍​‍​‍‌⁠⁠‌​