The ownership of macro assumptions in financial models is alarmingly inconsistent, and that inconsistency is a risk most organizations don't take seriously enough. Across industries, the answer to who decides what number goes into a rate or commodity price assumption ranges from the CFO to a treasury analyst to whoever built the Excel file and never faced a follow-up question. That range is not a sign of flexibility, it's a sign that the process is broken.
When a single assumption tied to a macro event can swing a valuation by millions, the fact that it sometimes comes down to "whatever felt right at the time" is not just informal; it's negligent. The problem isn't that different organizations use different roles. The problem is that most of those roles lack a formal mandate to document the reasoning behind the number. Without documentation, the assumption becomes a black box. When the model is reviewed six months later, nobody can reconstruct why the interest rate was set at 4.2% instead of 4.0%, or whether that commodity price accounted for a specific supply chain disruption. The model becomes a tool of memory rather than a tool of analysis.
What this means for anyone who builds or relies on financial models is that the assumption tab should never be an afterthought. It should be the most audited part of the spreadsheet. The person who owns the macro assumptions needs to own the documentation equally. That might be a dedicated role in a large treasury function, or it might be the analyst who built the model, but the expectation must be explicit: every assumption must have a source, a date, and a rationale. If the rationale is "this is what we used last year," that is a valid answer only if it is written down and flagged as a roll-forward assumption, not a fresh forecast.
The practical takeaway is straightforward: if your organization cannot name the single person responsible for each macro assumption and cannot produce a written justification for that number, then your financial model is less reliable than you think. Fixing that does not require a new department or expensive software. It requires treating assumptions with the same rigor as the formulas that depend on them. The next time someone asks who decides what number goes in, the answer should be a name and a file path, not a shrug.