Budget Season Politics: Getting Honest Numbers Out of Department Heads

Ask a department head to submit next year's budget and you will, with almost complete reliability, get back a number that is padded somewhere between ten and thirty percent above what they actually expect to spend. This is not usually dishonesty in the way that word implies malice. It's a rational response to a system that punishes accuracy and rewards padding, and understanding why department heads do this is the first step toward getting numbers you can actually plan against.

Why padding is the rational move

From a department head's seat, the incentives are lopsided. If they submit an honest, tight budget and something unexpected comes up mid-year — a vendor price increase, a hire that takes longer to backfill than planned, a project that runs over — they have to go back and ask for more, which reads internally as poor planning even when it isn't. If they submit a padded budget and finance trims it during review, they end up close to where the honest number would have landed anyway, but they've avoided ever having to ask for more later. Padding is a hedge against a system where asking for more mid-year costs more, politically, than being seen to have a little unused budget at year end.

The finance side, in turn, has learned that submitted budgets are inflated, so review meetings often apply a blanket haircut — cut every department's request by a flat percentage — which department heads have also learned to expect, so they pad further to compensate for the anticipated cut. Both sides are responding rationally to a game the other side is also playing, and the result is a negotiation dressed up as a planning exercise, where the actual information content of the submitted numbers is close to zero.

Ask for the plan before the number

The fix that works is not asking harder for honest numbers — it's changing what you're asking for first. Instead of opening with "what's your budget for next year," open with "what are you planning to do next year, and what does each thing require." A department head asked to justify a padded travel line item against a specific list of planned trips and events will produce a smaller, more defensible number than one asked to submit a total and defend it only if challenged. The plan-first sequence makes padding visible to the person doing the padding, not just to finance reviewing it later — it's harder to inflate a number when you've just written down the specific things it's supposed to cover.

This takes longer per department than a top-line budget request, which is exactly why most organizations skip it and default to the number-first version. The time cost is real but front-loaded — a plan-first budget conversation in September saves the far more expensive conversation in June about why a department is already sixty percent through a line item nobody can account for.

Reward accuracy, not just frugality

If the only budget outcome anyone notices is "came in under" versus "came in over," you are training every department head to submit padded numbers, because coming in under a padded number is easy and looks good, while coming in close to an honest, tight number carries real risk if anything goes sideways. Organizations that get more honest budgets over time tend to explicitly separate two different signals in their year-end review: how close was the forecast to actual spend, regardless of direction, and did spend come in under the number. A department that forecasts within a few percent of actual, even if slightly over, is giving finance more useful information than one that comes in twenty percent under a padded ask — and if only the second one gets praised, accuracy will keep losing to the appearance of frugality.

Give department heads a real contingency line

A meaningful share of padding exists because department heads have no other mechanism to cover the unexpected — no contingency line, no process for a fast mid-year adjustment, nothing except the padding they built in themselves. Giving departments a modest, explicitly labeled contingency allowance, separate from their operating budget, does something padding can't: it makes the safety margin visible and shared rather than hidden inside each department's own number. A department that knows a real, sanctioned buffer exists has less reason to inflate every line item defensively, because the thing padding was quietly doing — protecting against the unplanned — now has an honest home.

Read the pattern, not just the total

Once numbers come back, the most useful review isn't whether a department's total looks reasonable — it's whether specific line items match the department's actual recent spending pattern. A department requesting a flat fifteen percent increase across every category, with no category standing out as the reason, is the signature of a padded, undifferentiated ask rather than a considered plan. A department whose request tracks unevenly — up sharply in one category tied to a specific known change, flat or down elsewhere — is much more likely to reflect real thinking. Finance teams that review for this pattern, rather than just checking totals against last year, catch the padded submissions faster and can have the plan-first conversation with the specific departments that need it, instead of applying a blanket haircut that punishes the departments that were already being straight with the numbers.

Retire the blanket haircut

The habit that most reinforces padding is the one meant to correct for it: an across-the-board percentage cut applied to every submission regardless of how it was built. A blanket haircut punishes the department that submitted an honest, tightly reasoned number exactly as much as the one that padded generously, which means the department that did the work of building a defensible budget learns the same lesson as the one that didn't — that precision wasn't worth the effort. Replacing the blanket cut with a targeted review, informed by the plan-first conversations and the pattern-reading described above, takes more of finance's time in a given season, but it's the only version of the process that actually rewards the departments doing this well instead of quietly telling them not to bother next year.