Michael M Atkinson, Lindblom’s lament: Incrementalism and the persistent pull of the status quo, Policy and Society, Volume 30, Issue 1, February 2011, Pages 9–18, https://doi.org/10.1016/j.polsoc.2010.12.002
Navbar Search Filter Mobile Enter search term Search Navbar Search Filter Enter search term SearchCharles Lindblom never abandoned the incremental version of decision-making he introduced in 1959, but as his work progressed he increasingly lamented the impaired quality of inquiry that characterizes public (and private) decision-making. Lindblom did not identify the precise origins of socially created incompetence, but he made it clear that incrementalism is not the source. This article suggests that two lines of intellectual inquiry—one based on institutionalism, one on behavioral economics—provide persuasive accounts of the reasons for Lindblom’s lament. In each case the status quo is a central concept and its persistent hold over decision-makers is the reason for less than timely responses to policy deficiencies. Neither line of inquiry is inconsistent with incrementalism, but each improves on Lindblom’s original formula.
In 1959 Charles Lindblom produced a pioneering paper that purported to describe how decisions are made in complex organizations charged with creating and implementing public policy. His principal objective, other than providing a compelling picture of the policy making process, was to discredit an approach that he, and later Braybrooke and Lindblom (1963), labeled “synoptic.” Lindblom argued that the synoptic model, premised on a complete assessment of values, alternatives, costs and benefits, requires far too much information, far too much calculation, and far too much value consensus to be persuasive as a model of how decisions are made. Equally important, it urges a change agenda premised on central direction by enlightened utility maximizers. This, according to Lindblom, is not a recipe for policy success.
By 1979 Lindblom could claim, with some justification, that for most policy problems there is no alternative to disjointed incrementalism – a stratagem based on small steps, trial and error, and a limited consideration of consequences – as both a description of how policy is made and as a preferable strategy in the face of pervasive complexity. Synoptic analysis, an optimizing strategy based on perfect information and carefully weighted outcomes, was never a serious alternative except in the minds of some academic economists. The fact that Lindblom himself did not find incrementalism particularly controversial may suggest that his alternative was something of a straw man. But until Lindblom, Herbert Simon, and James March, decision theory was in thrall to hyper-rational approaches to organization, and nothing that Lindblom or his colleagues had to say about the limits to rationality diminished the advocacy of comprehensive decision-making methods, especially in the budgeting arena.
Despite Lindblom’s obvious commitment to incrementalism as a preferred decision-making strategy, in later writings he expressed misgivings about what he called the impaired quality of inquiry that characterizes public (and private) decision-making. Setting aside biological sources of impairment – our limited capacity to calculate and recall – in Inquiry and Change (1990) Lindblom expressed concern for socially created incompetence. Ironically perhaps, the father of incrementalism complained of the narrowing effects of socialization and social inequalities, the insidious effects of conformity, and the enervating effects of “political docility.” All of these have combined to produce responses to pressing problems that are seldom timely: “The train of enlightenment forever runs behind. In the best of circumstances the train never reverses or runs off the track and it makes all the station stops. But it arrives at each stop late” (1990, p. 70).
Not that Lindblom considered incrementalism to blame. The range of options under consideration at any given time is, Lindblom argues, narrower than even an incremental interpretation of the policy process would anticipate (1990, pp. 131–32). Why is that? What are the sources of conformity and narrowing? If it is not incrementalism that has reduced our horizons, what has? Lindblom struggles to answer this question and ends up with the usual suspects: media concentration, indoctrination, state information monopolies, and social inequalities. His solution (and it is a partial one because Lindblom is not sure that impairment is entirely without its advantages) lies in improving the competition among ideas to help spur “probing” and inquiry.
I want to argue, in agreement with Lindblom and others ( Weiss & Woodhouse, 1992), that incrementalism is not the source of our policy-making deficiencies; it is not inherently conservative nor does it imply policy drift. But I want to suggest that Lindblom’s lament and his diagnosis of the sources of impairment neglects a topic that he himself introduced, albeit tangentially, namely the role of the status quo in structuring our responses to proposals for policy changes. While Lindblom devoted his career to undermining arguments premised on the need to reoptimize every time there is a change in the policy environment, he spent very little time on the opposite, namely doing nothing at all. Incrementalism implies at least some degree of policy change, but what about no change? Could it be that Lindblom’s lament about our collective inability to be expansive, timely and innovative has something to do with the grip that the status quo exerts on us? Since his pioneering work, research agendas on political institutions and in cognitive psychology have gone well beyond “muddling through” in suggesting why the status quo is so important to understanding the process of policy change. They show, I believe, that the gap between Lindblom’s analysis and his lament is to be found in the persistent pull of the status quo.
Incrementalism is a quintessentially middle range theory. It is not a grand narrative that aspires to account for the entire policy process. But neither is it a collection of disconnected observations or even generalizations. Like most middle range theories ( Boudon, 1991) it seeks to account for a puzzling phenomenon by positing a set of subjective and objective conditions that are at its source. The phenomenon that puzzled Lindblom is why decision makers, in spite of exhortations to make decisions more holistically, routinely fail to do so. His answer is a theory that at once describes how decision-makers employ a method of “successive limited comparisons” and shows why this method works, both empirically and normatively.
Incrementalism’s core insight is easy to summarize. Decision-makers, who must respond to problems in the absence of certainty regarding outcomes or agreement over core values, will typically engage in a local search for options (“successive limited comparisons”). This search process results in small adjustments from the status quo premised on what is practical and what is possible. Practicality is an abiding criterion because disagreement over values is ubiquitous, even in the presence of prescribed objectives. Decision-makers end up choosing policies and values more or less simultaneously ( Lindblom, 1959, p. 82) rather than constructing and then implementing grand designs. Comprehensive alternatives are typically ignored because they are impractical in their political requirements or unpredictable in their consequences.
This core insight into decision-making dynamics has been embellished in a variety of ways, not least by Lindblom himself (1979), who offered additional important distinctions to overcome misunderstandings and to sharpen the applicability and persuasiveness of his original contribution. Most important, in work that built on his original article Lindblom placed incrementalism within the larger framework of what he called “partisan mutual adjustment.” This framework, hinted at in 1959, was explored more thoroughly six years later when Lindblom (1965) turned his attention to how incrementalism, now called “disjointed problem solving,” could sustain intelligent policies when practiced on a large scale. This shift to a macro level was one in which the limited capacities of individual participants continued to play a significant role, but now Lindblom endeavored to show that self-interested “partisans” practicing incrementalism could achieve policy coordination without having to impose a synoptic design. In fact, according to Lindblom, partisan mutual adjustment, the act of rationally adapting one’s options to the existing policies and the anticipated actions of others, is more likely to connect resources to collective preferences than any effort to accomplish this task via central command. The adjustments and adaptations that follow are not premised on conformity to a centrally defined common good, or even to its search. Partisans (and that term covers all players) have their own goals and seek to realize them within a constrained environment marked by uneven competition. Lindblom’s roots in economics are revealed most clearly at this macro stage where coordination by partisan mutual adjustment is analogous to coordination by markets.
With few exceptions Lindblom’s critics have also been ardent admirers. Rather than refute incrementalism, their main preoccupation has been to set some limits around it, making it a more genuine middle range theory. Lustick (1980) is a case in point. Incrementalism, he argues, works well in situations where the variables that decision-makers must take into account are continuous and there are no thresholds or sharp discontinuities. Similarly, where decisions cannot be broken into manageable pieces (e.g., large weapons procurements) or where budget requirements preclude trial and error, incrementalism is not always a sound strategy. In short, what Lustick calls the “task environment” is a much greater constraint on decision makers than Lindblom originally suggested.
To be fair, Lindblom was careful to insist that partisan mutual adjustment, for example, was not “useful” in all settings: “a general endorsement of it seems as misplaced as a general condemnation” (1965, p. 13). Similarly, the general superiority of disjointed incrementalism was not something that Lindblom was inclined to insist upon: it is always a matter of how well these strategies work “in various specific contexts” (1965, p. 293). That it was left to others to specify these contexts is no criticism of Lindblom, but it must be acknowledged that the original statement of incrementalism did not dwell on its limitations. Now, just as Lindblom suggested in 1979, it is standard to find incrementalism included as one of a series of potential strategies that are available to decision makers. Howlett and Ramesh (2003) for example, combine agency and structure to create a taxonomy of decision-making “styles”, only one of which is incrementalism. This style is best suited, they suggest, to situations in which the policy context is relatively simple, and the constraints on decision makers are relatively high. Under other circumstances, particularly those in which the problems are complex, decision makers often abandon incrementalism, not for a return to synoptic decision models, but for ones that acknowledge the presence of multiple actors whose choices cannot be counted on to achieve coordination via mutual adjustment. Howlett (2007, p. 663) makes the point, in this regard, that incrementalism becomes less relevant as a decision style when the context shifts from routine decision making in which some control of the environment is assumed, to policy making where a much larger measure of multi-level politics takes over.
The growing attention to context is one reason that Lindblom’s insights have become simultaneously commonplace (incrementalism is everywhere) and less pertinent (incremental responses are not necessarily the most interesting). As the environment for decision-making begins to acquire a stronger political frame, that is as divisions in society and differences in political power become part of the environment, “muddling through” begins to look like a rather limited strategy; not as limited in applicability as synoptic policy-making perhaps, but not invariably practical either. Lindblom himself recognized the need to expand beyond the model of a rather isolated individual decision maker, hemmed in by routine practice and available options, to encompass a world of pluralist competition in which bargaining and adjustment are vital.
That shift began almost immediately after his path breaking 1959 article and found expression in both A Strategy of Decision and The Intelligence of Democracy. In 1977 the publication of Politics and Markets signaled a shift toward an even broader environment, this one populated by groups that are structurally advantaged power blocs rather than simply mobilized interests. Lindblom’s thesis that business enjoys a unique level of political power in capitalist economies did not abandon the incrementalism formula for which he was already famous, but it did raise questions about its relevance. Lindblom (1979) recognized the apparent disjuncture and argued that no contradiction existed between his original formula and his newer claims. But the power dimension that Lindblom had originally downplayed was now in full view and incrementalism began to seem a very conservative strategy in the face of significant power differentials, a strategy quite congenial to those who sought nothing more than agreeable and manageable adjustments to the status quo.
As a result, incremental styles of decision-making, in addition to becoming only one of a number of alternative styles, have also become normatively less compelling. In a world of deep divisions between “haves” and “have-nots,” it is not obvious that the pressing problems of society should invariably be met by incremental strategies. Where information regarding the likely impacts of intervention is limited, and where implementation confronts a host of compliance challenges, incremental strategies continue to look good. But in the face of, for example, a collapse in the cod fishery, institutionalized discrimination, or the failure of the domestic automotive industry, it is hard to imagine a measured, incremental, response that is adequate. Adopting strategies that are familiar cannot be relied on to meet political needs or democratically defined expectations when entire belief systems are being challenged, problem definition is being contested, and cracks in social consensus are being revealed. Incrementalism, under these circumstances loses the normative purchase it had in more settled times.
In “Muddling Through” Lindblom offered an interpretation of the policy process from a hypothetical administrator’s point of view. His incremental model was a response to the question of how individuals perform in an environment characterized by significant levels of complexity. But as early as Politics, Economics and Welfare (1953) Lindblom, with Robert Dahl, had posited how incremental strategies at the individual level fit within broad processes of social control. The macro elements of Lindblom’s incrementalism continued to develop after “Muddling Through,” beginning with the robust pluralism of The Intelligence of Democracy and moving to a structural perspective that challenged conventional American political science ( Lindblom, 1982). At no time in this odyssey was the power of the status quo itself ever directly addressed, but Lindblom’s portrait of business’s privileged position in advanced capitalist societies led him to conclude that “corporate molding of volitions on grand issues is usually to confirm rather than to change volitions, since it extols the status quo” ( Lindblom, 1977, p. 207). He concluded Politics and Markets with the observation that the “pivotal” threat to polyarchy and the strategy of mutual adjustment is the proliferation of veto points within both government and the broader political system ( Lindblom, 1977, p. 347). It would be going too far to say that Lindblom anticipated the institutional turn in political science, or even that he adequately reconciled the agency formula of incrementalism to the structural formula of business advantage, but he was more aware than most of the need to unite the macro and the micro.
In what follows I describe how, despite major shifts in academic fashions, our understanding of the policy process remains tethered to Lindblom’s original insights while struggling to overcome some of their limitations. Recent scholarship on the policy process has not felt obliged to follow Lindblom and unite agency and structure, but it has placed more emphasis than Lindblom did on the status quo as a focal point for policy change and it does model, more effectively than Lindblom, the reasons for the absence of change in situations that seem to cry out for it.
As if to cement his argument about the ubiquity of incrementalism, Lindblom was at pains to avoid any reference to particular countries or political systems. He seemed to have very little use for what was once referred to as “comparative government.” In fact, in his original formulation Lindblom went beyond the admonition to substitute variables for the proper names of social systems in comparative analysis ( Przeworski & Teune, 1970, p. 25), and described his “muddling through” model without the help of any system-level variables as such. While he constantly invoked the American experience, the strong implication is that incrementalism is to be found everywhere, evidently without regard for institutional architecture or democratic culture. Even those who have followed Lindblom and placed boundaries around the applicability of incrementalism have not ventured far into the institutional realm.
Did Lindblom miss something important by declining to build system level variables into his model? The early stages of incrementalism outside of the American environment, particularly the study of budgeting, were promising for its long-term relevance to a wide range of political systems. However, as the study of policy change advanced, and institutional variables became systematically incorporated into research, incrementalism became a catchall term used to describe slow, uneven policy progress. Its relevance was not diminished, because slow, uneven progress characterized most policy change in most political systems. On the other hand, Lindblom provided no institutional interpretation of the conditions under which incrementalism might be in ascendance, and when it might be set aside for a bolder agenda. Whereas disjointed incrementalism implies a continuous process of small adjustments, institutional scholars interested in public policy have set themselves the task of identifying those enduring constellations of rules that both limit change and facilitate it. For them, the most compelling variables in policy analysis are the constitutional provisions and rules of political competition that structure the pace of policy change independent of inherent incremental predispositions.
Take, for instance, the veto player theory of policy change associated with George Tsebelis. This theory is useful to consider as an institutional and political complement to the incrementalist model because its focus is on the conditions that limit policy movement beyond the status quo. Interestingly, Lindblom decried what he called “political incrementalism” and added to the familiar dispersion of veto powers in the American political system his own concern for the veto powers possessed by corporate interests (1979, p. 520). But whereas the status quo featured in Lindblom’s analysis only as an annoying hindrance, a backdrop to policy change, in the more recent veto player literature, the status quo is the star. The theory argues that the more veto players that exist and the greater the ideological distances between them, the more entrenched the status quo becomes. While establishing the pattern and relevance of veto points is an imposing conceptual and empirical task, there are two general sources of vetoes: those that emanate from constitutional provisions, and those that arise out of partisan political arrangements.
Where constitutions provide for independent bases of support and authority for executives, legislatures and judiciaries, there are many veto points. In the United States, the presence of a strong upper house, federalism, and a presidential veto creates numerous institutional veto players. These are enhanced, in theory at least, to the degree that control of strategic points is shared among competing partisans rather than concentrated in a single party. Parliamentary systems either have single veto points, as in the case of majority governments, or multiple and complicated veto points depending on the ideological make-up of the governing coalition. The more parties that are required to form a government, the more veto points; the farther apart these parties are in ideological terms, the smaller the “status quo winset,” that is the policy positions to which all players are willing to move.
The theory’s implications for policy change are remarkably simple: “countries with many veto players will engage in only incremental policy changes” ( Tsebelis, 2000, p. 464). Significant policy change is more difficult to achieve as the number of veto players increases. The question of what “significant” means is naturally an issue, but Tsebelis and his colleagues are persuaded that it is possible to identify changes that qualify by relying on external expert assessment. With a measure of “significant” legislation in hand, here is how Tsebelis (1999, p. 605) describes the conditions for policy change in parliamentary government: “If there are many veto players separated by large ideological distance, then legislation can only be incremental.” Little or no significant legislation can be passed and a crisis of governability eventually results. In parliamentary systems this crisis manifests itself in a series of coalitions that come and go without much legislative effect; in presidential systems pressure is placed on the regime itself. Where the judiciary has veto powers, stalemate provides an opportunity for the expansion of judicial authority: “the court may interpret statutes any way it wants within the Pareto set of the veto players without fear of being overruled” ( Tsebelis, 2000, p. 466).
How does this stack up against Lindblom? It is first of all a political thesis. The emphasis is on policy shifts associated with politicians who, in turn, are responding to the interests of constituents. Bureaucrats play only opportunistic roles, not creative ones. Bureaucratic behavior is a function of the distribution of veto players: the more of these, the more likely legislation will be cumbersome and complicated in an effort to restrict discretion ( Tsebelis, 2002, pp. 236–237). Second, the politicians do not resemble the encumbered strategists that incrementalism presumes. They are, in Bawn’s words, “fully responsive” to their strategic environment (1999, p. 709). This is important because the veto player model does not anticipate gradualism. If policy alternatives lie outside the current Pareto set for existing veto players, no change will occur; if they lie inside, change will be “significant.”
Notwithstanding the empirical support the veto player version of institutionalism enjoys, its decision-making theory, to the degree one is articulated, is rather deterministic and mechanical. The micro level assumptions are firmly in the rational choice tradition with decision-makers unencumbered by the cognitive limitations on which Lindblom (and many others) focus. In the veto player model policy assumes a stop-and-go quality with a heavy emphasis on stop. There is no allowance for mutual adjustment beyond the institutional core, even though research shows that governments, hemmed in by coalition partners, constitutional veto players, and the tyranny of the electoral cycle, can be both pragmatic and imaginative. Periods of institutional stasis, Thelen (2003, pp. 225–226) points out, are often accompanied by subterranean change with governments adjusting the roles of existing institutions, adding new ones, and reaching accommodations with other governments and societal partners. Incremental change, the only kind of change anticipated in the presence of strong veto players, can nonetheless produce significant policy shifts over time ( Coleman, Skogstad, & Atkinson, 1996; Howlett & Cashore, 2009, pp. 41–42996). Lindblom (1979, p. 521) put it this way: “Incremental changes add up; often more happens than meets the eye.” It is even possible to imagine veto players accepting short-term sacrifices to their policy interests in order to maintain long-term coalition success ( Scharpf, 2000, p. 782). In short, an institutional analysis built so heavily on the status quo may underestimate the possibility of policy change achieved in increments.
Still, the force of the institutional argument remains. Where single veto players dominate, policy change can be dramatic and rapid; on the other hand, a pluralistic situation, with multiple actors and a variety of veto points, is less congenial to coherent and timely policy choices. Lindblom (1977, pp. 346–348); Lindblom (1979, pp. 520–521) laments the proliferation of veto points, but he makes no attempt to incorporate systematically the effect of institutional context on the practice of incrementalism. Institutional theories emphasize impediments to decision-making and highlight the need for dialogue and compromise, but Lindblom does not introduce institutional variables into his model. On the contrary, he is at pains to distinguish incremental politics, presumably practiced within political institutions, from incremental analysis, presumably practiced in the depths of bureaucratic organizations. When he does consider institutions it is to dismiss them as sources of creative problem solving: “no society or subgroup within it ever designed a state or government to make it an effective instrument for social problem solving” (1990, p. 286).
Political institutions may be insufficiently supple to react well to changing policy environments, and they may not be ideal places to practice incremental analysis. That is not, however, a sufficient reason to ignore them. And in ignoring them Lindblom also ignores some of the sources of his own lament. If the train of policy change arrives too late it is sometimes because the institutional conditions are not suited to problem solving on time. Incrementalism, in this reading, is neither an impediment nor an asset; it is merely the by-product of an institutionally orchestrated attachment to the status quo.
Incrementalism is a theory of organizational behavior. It argues that confronted with deficiencies in current policy, the most likely, and reasonable, organizational response is minor adjustment to what already exists. In Lindblom’s hands, nothing could be more rational than reducing risk by engaging in trial and error.
As we have seen, Lindblom’s early critics made the point that the rationality and efficacy of this kind of strategy is deeply contextual. Incrementalism may be rational, but that does not invariably make it the most appropriate response. The behavioral economics movement goes farther by suggesting that strategies that are recognizably incremental may nonetheless be seriously flawed because mistakes are often made in assessing the context of decision-making. Some of the most important insights have come at the hands of Kahneman and Tversky (1979) whose prospect theory, an account of decision making under risk, argues that how we frame information influences our judgment and our declines to stipulate the circumstances that favor incrementalism, Kahneman and Tversky devote themselves almost entirely to mapping the conditions under which people systematically violate the axioms of expected utility theory. And where Lindblom presumes that incrementalism is a rational response to decision contexts, Kahneman and Tversky paint a picture of decision-making highly susceptible to irrational impulses.
Lindblom’s portrayal of the decision-maker’s world emphasized complexity; prospect theory emphasizes uncertainty and the difficulties that people confront in estimating the likelihood of uncertain events. The reduction of risk is a central preoccupation of both theories, but rather than relying on the simple expedient of incremental steps as a solution, prospect theory argues that decision-makers deploy a limited set of heuristic principles “which reduce the complex tasks of assessing probabilities and predicting values to simpler judgmental operations” ( Tversky & Kahneman, 1974). These heuristics, which include, for example, the inclination to look for similarities in decision situations, can be extremely useful ways of coping with uncertainty. If a new, unregulated nutritional therapy appears on the market, consider regulation that is identical to the regulations used for other, similar, products. Lindblom would presumably agree. However, as Kahneman and Tversky make clear, these heuristics are also sources of systematic bias and it is the emphasis on biased decision-making that distinguishes prospect theory from incrementalism. Whereas Lindblom was content to stipulate the cognitive limitations of decision-makers and advocate incrementalism as a response, prospect theory explores the character of these limitations and shows us how the biases they generate can undermine any response, including incremental ones.
Prospect theory’s most important contribution to the behavioral economics movement is the observation that people behave differently in confronting decision situations in which outcomes are framed as losses compared to gains. Numerous experimental studies have determined that most people are loss averse, that is they are less willing to risk losing what they already have than to risk gaining an equal amount. Most respondents will not stake $10 on a coin toss if they stand to win less than $30 ( Kahneman & Tversky, 1984). Another way to put it is that in the realm of losses, people are risk seeking; they are willing to take chances to restore or prevent losses. In the realm of gains, people are risk averse; they are inclined to choose a sure gain over a positive gamble. In conventional, neo-classical decision analysis, these differences should not be observed. The focus should be on total wealth or final outcomes, not assets framed in terms of gains or losses.
The observation that people would often demand more to give up an object than to acquire it has been labeled the “endowment effect” or the “status quo bias.” The endowment effect implies that the value attached to an item increases the moment it is in one’s possession. A status quo bias is the presumed result: individuals are disinclined to give up items with which they are familiar and to which they are attached independent of their market value. And while it may be dangerous to extrapolate too far from the laboratory, the tendency to follow customary policy, re-elect incumbents, purchase that same products and stay in the same job are all behaviors that speak to our preference for the status quo ( Samuelson & Zeckhauser, 1988, p. 8). The inclination to evaluate alternatives using a reference point is a common trait, but it does not fit well with the classical decision-making model in which outcomes should not be influenced by factors that are preference-irrelevant, such as irrational attachments to the status quo.
Of course, attachment to the status quo is not always irrational. Often transition costs are very high and there are positive returns to continued investment in particular technologies or decision strategies ( Pierson, 2000). It is not easy to contemplate switching from analog to digital when the switch involves abandoning huge investments. And while changes of this magnitude occur because of the demonstrable benefits, it is not always clear that change is change for the better. Uncertainty also plays a role in discouraging change with or without endowment effects. But the latter make an independent contribution to status quo bias when choices are framed as losses. Thus decision-makers whose elaborate and expensive programs do not generate the expected capital investment, the anticipated immigration of skilled workers, or the desired amelioration of educational deficiencies, cannot be counted on the make the necessary (even incremental) adjustments. The idea that decisions should be based on incremental costs and benefits is swamped by the even more powerful intuition, that a large historical resource investment must be recouped or otherwise justified ( Thaler, 1980).
Do these findings, which are the tip of a monumental iceberg, somehow undermine Lindblom’s incremental insights? Yes and no. They focus attention on the important role of incremental gains and losses in the minds of decision-makers and, in that sense, they confirm that increased uncertainty prompts modest, reliable change strategies. The potential gains from large, even revolutionary, steps are not foremost in the minds of most people confronting an uncertain future. On the other hand, to the extent that the heuristics used to cope with these situations introduce systematic biases, such as the status quo bias, then Lindblom’s argument that incrementalism is a superior strategy in most circumstances begins to lose ground. It is not that small, incremental steps are irrational, but decision-makers may refuse to take even those steps. Increments are steps away from something that already exists. What Lindblom has underestimated is that attachment might discourage the taking of any steps whatsoever.
This tendency to cleave to what is known and resist reform can be aggravated in a number of ways. Consider these three possibilities drawn from behavioral economics. First, changes in policy will often generate losses for some and gains for others. Where the distribution of losses and gains is unclear, research suggests that efficiency-enhancing reforms are less likely to be adopted, even if, in retrospect, they would have proven popular ( Fernandez & Rodrik, 1991). Comprehensive trade agreements are a case in point. Sector or commodity based agreements, consistent with incrementalism, are easier to negotiate and tailor to political considerations. Comprehensive agreements, on the other hand, produce larger aggregate benefits but create individual-level uncertainty and thus resistance to change. In the case of NAFTA, for example, ex ante hostility turned into ex poste support, but it took a major policy step, one that was not consistent with disjointed incrementalism, to achieve this end.
The status quo effect is further compounded by an emotional tendency to prefer negative outcomes caused by inaction to negative outcomes caused by action. For example, many parents prefer not to vaccinate their children when the risk of death from disease is 10 in 10,000 and the risk of death from vaccination is less than 5 in 10,000. In this case preference for the status quo (no vaccination) is bound up with, and exacerbated by, a preference for inaction. Known as omission bias this tendency has been attributed to the belief that outcomes caused by omissions are natural and carry with them a reduced degree of personal responsibility ( Spranca, Minsk, & Baron, 1991). Put another way, omissions have less causal import: “omissions…tend to be evaluated as ‘neutral’ regardless of their outcome, while commissions are evaluated as positive if their outcomes are better and negative if their outcomes are worse than the presumed outcome of inaction” ( Ritov & Baron, 1992, p. 60). Lindblom’s account of incremental change presumes some form of action on the part of decision-makers. Omission bias, however, is a reminder that even small changes that require effort and commitment are subject to framing illusions (omissions versus commissions; losses versus gains) that make them appear less attractive than the status quo.
Finally, while omission bias does not include a requirement of accountability, most people feel accountable in organizational decision settings. Lindblom never spoke about who would be responsible for incremental decisions. His is a model in which decision-makers have no necessary relationship to one another or to a more remote principal. And yet, politics is about these relationships. While Lindblom did not address accountability directly, it seems reasonable to argue that incremental decisions, to the extent that they reduce risk, may be particularly attractive to those who are obliged to explain and justify them. But a heightened sense of accountability may also magnify the status quo effect. Tetlock and Boettger (1994), for example, have shown that political accountability demands encourage decision makers to weigh more heavily lives lost from changing the status quo than lives lost by preserving it. In deciding whether or not to agree to the release of a new drug, experimental subjects who were required to be accountable resisted placing a drug on the market if it might have lethal side effects, regardless of whether it also had a favorable benefit-cost ratio. “In effect, subjects were willing to allow more people to die than to live” ( Tetlock & Boettger, 1994, p. 19). In the case of broader policy initiatives, the requirements of accountability may have similar effects; abandoning the status quo may look relatively unattractive if the results can be clearly traced to a specific decision-maker.
The work done under the rubric of behavioral economics is broadly consistent with incremental decision-making. In particular, Lindblom shares with behavioral economists a suspicion of the canonical decision making model with its emphasis on comprehensive (synoptic) calculations of costs and benefits. The major disjunction concerns the reference point from which decisions are made. For behavioral economists, like veto player theorists, the reference point is fundamental to understanding subsequent outcomes, and one of the most important of these reference points is the status quo. Lindblom’s decision makers are located in a status quo as well, but for Lindblom the status quo does not have the same capability to distort decisions. It is the starting point from which changes are made, but change is presumed to be a necessity or a constant condition. The concept of status quo bias introduces the very real likelihood that decision-makers will avoid any kind of action at all. The result is that while Lindblom invites us to be positive about incrementalism, prospect theory invites us to anticipate mistakes in incremental calculations, especially where the decision frames are capable of manipulation ( Mercer, 2005).
It is no criticism of Lindblom to say that other, more comprehensive perspectives (e.g. Jones & Baumgartner, 2005) have absorbed his theory of the policy process and contextualized his acute observations. Given the speed with which academic fashions change, it is no mean achievement that the language of incrementalism persists as a commonplace in most accounts of the policy process. Mind you, no one is prepared to argue any longer that incremental strategies are invariably the best. The fact that Lindblom himself guarded against that claim does little to mitigate the implied criticism since he spent very little time theorizing incrementalism’s applicability. The great power of incrementalism, its implied universality, turns out to be its great weakness. But that was to be expected. Lindblom set the stage for decades of research and comment; the result is a much more refined idea of when small or radical policy change is likely to take place.
Lindblom’s major achievement lies not in coining terms like “muddling through” or attacking rational decision making models; it lies instead in anticipating the heuristics movement represented in contemporary social decision theory. Lindblom rooted his theory of decision-making in an appreciation of cognitive shortcomings and the ways in which decision-makers compensate. His heuristic was the idea of a local search for policy alternatives. What Lindblom did not anticipate well enough was that heuristic shortcuts bring biases with them ( Bendor, 1995, p. 833). Lindblom was well aware of the status quo, for example, but not of status quo bias. His major concern, that decision makers would be induced to launch a comprehensive search for risky new alternatives, was never complemented by an equal concern that decision-makers might reject any change whatsoever.
Yet in a rather disjointed (pun intended) fashion Lindblom did identify the three principal reasons why the status quo has such a persistent pull. First, Lindblom emphasized throughout his work the debilitating effects that inequality has on the prospects for political change. In his political economy the marriage of political and economic power removes the volition for change and grants the status quo an enormous advantage over alternatives that could threaten established patterns of dominance. Second, although he was unwilling to build institutional variables into his decision-making models, Lindblom consistently lamented the proliferation of veto points, institutionally generated opportunities to thwart change, incremental or otherwise. Finally, Lindblom saw clearly that cognitive constraints, combined with complex problems and many decision-makers pursuing inconsistent goals, stand squarely in the way of comprehensive problem solving. The psychological work that followed Lindblom’s original observations shifted the emphasis from complexity to uncertainty and revealed, in ways that Lindblom could not have anticipated, the power of the status quo as a starting point for all decision making. In short, there is more continuity than discontinuity between Lindblom and the institutional and behavioral approaches that followed his path breaking work and he succeeded, where others barely try, to connect micro and macro in a broad portrayal of the policy process.
But Lindblom’s lament is still with us. I have argued that dissatisfaction with the quality of social decision-making has its roots in the strategic advantages enjoyed by the status quo. The fact that Lindblom may have underestimated or even misunderstood its effects should not overshadow the role that incrementalism can play in overcoming status quo bias. After all, the incrementalist creed urges us to experiment and the conditions for experimentation are strongest in the open societies on which Lindblom concentrated. Trial and error, as Woodhouse and Collingridge (1993) have emphasized, is both the heart of the incrementalist strategy and in all likelihood the most effective way of enticing decision-makers to contemplate a better future. It will not work on all occasions, especially where the feedback loops are long and the risks of experimentation are high. But if we are to avoid Lindblom’s lament, that the train of enlightenment persistently runs behind schedule, then we need theories of policy change, including incrementalism, that take the status quo seriously and work in determined ways to remove its overweening influence.