#1279 1969 · Texas Instruments (Peter Pyhrr) · Corporate finance & budgeting

Pyhrr made departments rebuild budgets from zero instead of adjusting last year's number

the problem

Corporate budgets renewed themselves automatically each year because no one ever had to re-justify old spending

background

The standard corporate budgeting process at Texas Instruments, like nearly every large organization, started with the prior year's approved budget and negotiated adjustments at the margin — a 3% cut here, a 5% increase there — with departments defending only the change, never the underlying spending itself. That structure meant that once an activity was funded once, it tended to stay funded indefinitely regardless of whether it still made sense, because no step in the process ever forced anyone to answer the question 'would we fund this today if we were starting from nothing?'

Peter Pyhrr, a manager at Texas Instruments, saw that incremental budgeting made waste self-perpetuating: obsolete programs survived year after year not because anyone actively defended them, but because the baseline itself was never re-examined, only adjusted. Cutting waste through the normal process meant hunting for it inside thousands of line items that were all, by design, treated as already justified.

what everyone would do

The standard cost-cutting response was to apply an across-the-board percentage cut to every department's existing budget, or to have finance hunt for waste inside a budget structure that still treated last year's spending as the presumptively correct baseline — both of which leave the underlying, unexamined spending largely intact.

what they saw

Last year's budget is not neutral — it's an unexamined pile of every prior year's spending. Deleting that baseline, forcing every line to re-earn funding from zero, exposes what nobody could defend today.

the move

Pyhrr discarded the prior year's budget as a starting point entirely and required every department to build its request backward from zero, breaking each activity into 'decision packages' that had to independently justify their cost against the benefit they produced, ranked against every other package company-wide as if none of them had ever been funded before — reversing budgeting's normal direction from 'adjust last year's number' to 'start at nothing and earn every dollar back.'

why it works

Incremental budgeting only ever tests the marginal change, so any spending baked in before the current review cycle is structurally protected from scrutiny no matter how obsolete it becomes. Rebuilding every request from zero removes that protection entirely: every dollar has to compete on its current merits against every other dollar company-wide, not just against its own prior-year version. This surfaces spending that would never have been approved if proposed fresh today, precisely because it was never actually re-evaluated, only carried forward.

the payoff

Adopted at TI in the late 1960s, then used by Georgia under Carter in 1973, reporting about $55 million in first-year savings.

where it breaks

It is expensive and time-consuming to execute properly — building and ranking decision packages for every activity in a large organization is a major undertaking that can't be repeated every year without exhausting management, which is why most organizations that adopt it use it periodically rather than annually. It can also be gamed if departments learn to write justifications that sound compelling regardless of actual merit, and it risks disrupting genuinely necessary but hard-to-quantify functions that don't package neatly into a cost-benefit case.

what came after

Carter brought zero-based budgeting to the federal government after becoming president, and the technique became a recurring tool corporations and governments revisit during cost-cutting cycles, cited by McKinsey and others as a periodic corrective whenever incremental budgeting has let unexamined spending accumulate for too long.

references

[1]Zero-Base Budgeting in Historical and Political Context: Institutionalizing An Old ProposalPublic Productivity Review (SAGE), 1977journals.sagepub.com