Incrementalism as an Ethic

What Budgeting Actually Is

Reading a budget is hard, but not for the reason most people assume. The mechanics of the thing — the columns, the line items, the forecasts — can be confusing, but they're learnable. The real difficulty is interpretive. Why was this chosen over that? What reasoning sits behind a projection, an expenditure, a bet placed on next year? Once you start asking those questions, budgeting stops looking like accounting and starts looking like argument.

That shift matters most in the public sector. For-profit entities are profit-directed; they have a scoreboard. Government and nonprofit entities are mission-directed, and a mission is a contested thing. Every stakeholder around a public budget is competing with the others — over what the mission even means, over their own roles and responsibilities, over their capacity to hit their goals, and over a finite pool of resources. There is no single number that resolves the fight.

This is why I find it useful to keep the layperson in view. A descriptive budget theory has to start from the honest premise that information is unevenly distributed. Nobody sees the whole board. Even the most tenured, most skilled bureaucrat in the room is still working from a partial view, still susceptible to the biases of their own vantage point. Any framework that pretends otherwise is describing a world that doesn't exist. And that is the door through which incrementalism enters.

Incrementalism as Budget Theory

Incrementalism, as a descriptive and positive model, holds something almost deflating: budgetary decision-making is largely a matter of building from where you already are. Across most time horizons, very little actually changes. Last year's figure is this year's starting point, adjusted at the margin. The theory doesn't treat that as a failure of nerve. It treats it as a rational response to conditions — because the players and their agencies are in constant conflict with one another over influence and over larger shares of the pool.

The intellectual scaffolding here is bounded rationality. As Daniel Williams and Thad Calabrese put it in their work toward a metatheory of budgeting, Herbert Simon argued that decision-makers simply don't have the resources — above all, the time — to make the comprehensive rational calculations that classical economic theory demands of them. Charles Lindblom pushed the point further: it isn't just that we lack time, it's that decision-makers aren't smart enough. When they rely on theories about what will happen without the experience to back those theories up, they are likely to be wrong.¹ Read that way, small, corrigible steps aren't laziness. They're error-correction under constraint.

So I'd define incrementalism this way: it's the framework in which the chief measure of effective budgeting is the ability to make a decision responsive to current constraints, right now. That definition does real work, because it deliberately strips deliberation out of the budgetary process. And here Aaron Wildavsky's point in The Politics of the Budgetary Process is essential. The very idea of "employing strategies" in budgeting unsettles people; they want programs presented on their merits alone. But that wish, Wildavsky says, presupposes agreement on what merit is — when the actual problem is that people don't agree. That disagreement is precisely why we have politics. To lay down and enforce fixed criteria of merit inside the budget would be to decide in advance what government shall do, and thereby to abolish the need for politics altogether.²

Read that way, incrementalism as budget theory shades naturally into a policy posture: advocate for as little change as possible. Agencies and their bureaucratic hierarchies exist to codify and stabilize processes against a churning world. You could put it more sharply — bureaucracy exists to subvert change through the sheer inertia of its stabilizing routines. Anyone who has read Durkheim or Weber recognizes the modern bureaucracy as an engine built to stave off change through efficiency and scale. Budgeting is no different. Its incrementalism is the fiscal expression of that same instinct.

Then as Policy: The Illinois Warning

If budgeting is the mechanism through which policy is actually enacted, then the vast menu of competing budget theories — some rational, some positive, some prescriptive, some merely descriptive — is itself evidence of how contested the act of allocating money really is. You can see it just by comparing states. Budget reporting differs from place to place because priorities and practices differ from place to place.

Illinois is the cautionary version of that comparison. When you look at the state's audit reporting against a state like New York, you see what happens when policy consideration drops out of the budgeting process: structurally unsound practices that eventually destabilize the whole thing. Illinois's audit record shows improper accounting and reporting that left roughly half of all state expenditures out of the General Fund's spending reports — and shows a state trying to paper over the gap through bond purchasing and refinancing that never actually reckoned with its structural deficits. When Governor Blagojevich expanded spending and public-entitlement benefits without offsetting the increase, the state's already precarious footing got worse — and this in a state that made extremely little use of taxing taxable services in the first place. Incrementalism, taken as license to just keep building on the base, will happily build you straight off a cliff.

There's a deeper problem lurking underneath the Illinois case, and it's a definitional one. Budget theory has never settled on what a budget actually is. The question that has organized seventy years of the field may itself be the wrong question. As the metatheory work argues, V.O. Key's famous question wasn't really about budgeting at all — it was about appropriating, an older legislative act that budgeting was meant to replace. If Key confused appropriating with budgeting, then much of what we call budget theory is really a theory of appropriation wearing the wrong name.³ The history is real: modern budgeting takes its contemporary shape in the aftermath of World War I, after a long stretch of states reporting wild swings of deficit and surplus. But a genesis isn't a definition. Budgeting still can't cleanly say what it is — and the largest implication of that failure is that it can't distinguish itself from politics. My own read is that this isn't a bug to be fixed. Budgeting can't be separated from politics because budgeting is part of the political process.

The Top-Down Blind Spot

Here is where I think incrementalism, for all its realism, has a genuine flaw. As a political framework it's almost Confucian — preoccupied with professionalism and the longevity of departments, presuming a rigid orthodoxy in state agencies that outlast the executives who nominally command them. That presumption isn't baseless. But it leads the theory to look in only one direction.

Ever since the Budget and Accounting Act of 1921, budgeting has had top-down elements as well as bottom-up ones. The trouble is that the top-down process is less routine, less publicized, and involves fewer people, so it's poorly documented. What we know about it comes mostly from insiders' accounts and a handful of careful historical analyses. As Bozeman and Straussman describe it, budget theorists have had only limited success with the top-down side because there's little conveniently obtained data and little routine to observe — so most theorists simply don't try. Top-down features get ignored or written off as historical "disturbances." And yet those top-down elements are often as decisive as the hundreds of disaggregated, visible negotiations between OMB, the agencies, and congressional subcommittees that make up the bottom-up process.⁴

They go further, calling this a major omission: both processes matter in every budget cycle, and their relative weight shifts cycle to cycle with fiscal policy, presidential leadership, economic growth, foreign conflict, congressional assertiveness, and more.⁵ I think that indictment lands. A positive, descriptive model earns its keep by accounting for the real-world conditions acting on every player. If incrementalism systematically underweights executive agenda-setting — the place where the biggest allocative moves actually originate — then it's describing half the board and calling it the whole game. And because budgeting can't be pried loose from politics, that omission isn't a technicality. It compromises the integrity of the framework itself. I'd name the mechanism plainly: this is anchoring bias built into a discipline. The bottom-up process generates mountains of data, so the analysis anchors there; the top-down process generates little, so it drops out. The available information shapes the structure of the analysis, and the unavailable information quietly disappears.

Now as Ethic

So the harder question is whether incrementalism should be adopted as an ethic — whether an elected leader should take it on as a personal disposition or a governing platform, and not merely observe it as a pattern in the data. The appeal is obvious. Framing problems as solvable with achievable solutions makes both the achievement and its payoff concrete. Fixing an immediate problem produces a visible result. And holding the status quo is, most of the time, treated as a good in itself.

But the power of the purse cuts both ways. If a policy will do more long-term harm than good, then obstructing the money that feeds it is the responsible move, not an abdication. Consider a mundane version: a department gets a healthy budget for technology upgrades and spends it buying twenty-plus new computers with their Windows licenses — when a single new server, next-generation chipset and modern OS architecture, could have virtualized all twenty-plus employees through thin clients. The server route may well be cheaper and far more upgradeable across a five-year horizon, but that option never surfaced at the proposal stage because the information wasn't in the room yet. That's incrementalism failing in miniature: the base got extended, a decision got made, and the better decision was never even considered. Here, separating budgeting from politics — from real deliberation — is a disservice to everyone the budget is supposed to serve.

And yet I won't discount the framework wholly, because its core virtue is real. Without incrementalism's ability to reduce the political and procedural variables, legitimate action in the public sphere would drown under the sheer weight of competing agendas. Getting many players to move together toward a shared goal is genuinely hard; any long-term strategy aimed at significant change has to reckon with the barriers, and even with majority support behind it, competing interests and institutional friction force compromise and sacrifice.

Health care is the case that holds both truths at once. In 1974 President Nixon proposed state exchanges where people could buy insurance, while Senator Ted Kennedy, alongside Rep. Wilbur Mills, pushed a near-universal national insurance option. Both men wanted universal coverage. But rather than build on Nixon's market-oriented design, Kennedy pressed the public-oriented route, extending the logic of Medicare and Medicaid — programs that, for all their broad support, represented a break from the traditional government posture on health care. Had Kennedy chosen instead to work with Nixon, the result might have been a more moderated, genuinely incremental step toward the shared goal. Instead, meaningful insurance expansion didn't arrive until the Affordable Care Act in 2010. And here's the irony that ties the whole argument together: the ACA can be read as anti-incremental, sharply expanding state involvement in insurance against the deregulatory current running since the Reagan 1980s — yet it was still built on the existing Medicaid and market scaffolding, an incremental structure carrying a revolutionary payload. That's the lesson. Even landmark reform is assembled incrementally on what already exists, and top-down considerations shape the incremental movements — budgetary and political alike — far more than the bottom-up story admits.

So I'd keep incrementalism, but I'd keep it honest about itself. As a descriptive model it earns its place by taking information asymmetry, competing agendas, and the endless jostling for larger shares seriously. Its virtue is action — deciding, and thereby carrying out the political bargains already struck. It rightly wants to keep budgeting distinct from politics, so that the deliberation stays in the deliberative arena where it belongs. But it de-emphasizes executive agenda-setting almost to the point of erasure, and it rests on a discipline that still can't define its own object. Budgeting keeps failing to separate itself from political deliberation for the simplest of reasons: it never was separate. It is politics, denominated in dollars.