Drift and Conversion: Hidden Faces of Institutional Change
Jacob S. Hacker (Yale)
jacob.hacker@yale.edu
Paul Pierson (U.C. Berkeley)
pierson@berkeley.edu
Kathleen Thelen (MIT)
kthelen@mit.edu
Prepared for the Annual Meetings of the American Political Science Association,
Chicago, IL, August 30, 2013.
We would like to thank all the participants in the MIT Conference on Comparative Historical
Analysis in the Social Sciences for helpful comments. James Mahoney deserves special thanks
for his continuing thoughtful reactions.
Abstract
In advanced industrial societies, institutions and policies that rest on state enforcement
power change over time—often fundamentally. Scholars (and citizens) pay a great deal of
attention to these changes when they follow the normal script: democratic political bodies enact
new formal rules. Yet we know far less about how institutions and policies change without such
formal revision. In this paper, we examine two ways in which this happens, “drift” and
“conversion,” and their implications for institutional theory and political life. Drift occurs when
formal rules are deliberately held constant in the face of major environmental shifts, causing
their outcomes to change. Conversion occurs when political actors reinterpret ambiguous rules or
use the discretion inherent in them to redirect them toward new purposes. We consider several
features of institutions and their context that make each of these processes more likely,
particularly the automaticity and discretion embodied in formal rules and the character of the
social environment. In neglecting these two modes of change, we argue, students of politics have
overstated institutional stability and missed important processes through which a dominant
political coalition’s victory may erode. They have also missed a significant source of potential
influence for opponents of existing institutions or policies, who can, through drift or conversion,
transform themselves from embattled enemies of the status quo into politically advantaged
defenders. Most notably, these often-hidden processes provide a means by which political actors
can alter governance without bearing responsibility with the broader public. They thus brings
into view the sometimes limited and conditional character of voter influence while highlighting
the central role of organized interest groups with the resources, information, and foresight to
influence policymakers’ less-visible decisions over the long term.
Change has always posed knotty problems for social scientists (Huntington 1971). It
poses a particular challenge to institutionally oriented scholars who study long-term processes.
Unlike researchers focused on behavioral or strategic dynamics over short periods,
institutionalists with extended time horizons frequently find themselves seeking to explain major
changes in how institutions operate or affect society that are not easily reducible to short-term
shifts in preferences or coalitions. Yet for many years, the theoretical tools of comparativehistorical analysis were weakest precisely where the advantages of the method were greatest:
explaining the major institutional transformations seen over the longue durée.
That is no longer true. Comparative-historical institutionalists have not only relaxed
earlier assumptions about the fixity of institutions and exogeneity of institutional change; they
have also started to develop theories that explain precisely what researchers with a long-term
perspective so often observe: that stability in formal institutions can coexist with major changes
in those institutions’ purposes and effects. Nonetheless, scholars still lack a clear understanding
of the circumstances, processes, and strategies that produce such fundamental changes without
formal revision. It is these sorts of change—which we call “hidden,” because they occur despite
surface continuity in institutional rules—that are the subject of this paper.
We focus, in particular, on two forms of change that are frequently obscured from the
view of analysts (as well as citizens) but are also ubiquitous and often highly consequential:
“drift” and “conversion.” Drift occurs when institutions or policies are deliberately held in place
while their context shifts, changing their effects. A simple example is the U.S. minimum wage,
which is not indexed to inflation and thus declines in value unless new federal legislation is
enacted. In cases such as this, what is triggering change can be referred to as “context
1
discontinuity”—in this case, inflation—so that those wishing to effect change need only block
the updating of existing rules in the face of environmental change. Drift thus depends on how
sensitive the effects of an institution or policy are on its context and on whether policies are
designed in ways that foster their updating in the face of changing circumstances.
Conversion, by contrast, occurs when political actors are able to redirect institutions or
policies to new ends—that is, use them for purposes beyond their original intent. An example is
the Sherman Antitrust Act of 1890, which prohibited organizational forms that were found to be
operating “in restraint of trade.” Passed in a context of widespread concern about corporate
collusion and cartels, the explicit target of the legislation was business trusts. Yet even as large
corporations were finding ways around the law, they worked through the courts to powerfully
redirect the legislation in their battles against trade unions whose organizing efforts were then
similarly found to be “in restraint of trade.” In this case, what triggers change is “actor
discontinuity,” as political actors who were not part of the majority coalition behind a rule or
policy nonetheless find a way to use these rules to achieve their own (sometimes very different)
goals. Conversion thus feeds off of rule ambiguity and the multiplicity of political arenas in
which ambiguous rules can be reinterpreted by those seeking to use them for new purposes.
In prior writings, we have explored drift (Hacker 2005; Hacker and Pierson 2010, 2011)
and conversion (Thelen 2004; Streeck and Thelen 2005) largely as separate processes. Our
collaboration here represents an attempt to present a unified perspective that draws on a new
generation of research motivated by unease with deterministic varieties of institutionalism that
imply we can read outcomes off institutional rules. These efforts have resulted in a variety of
arguments that overcome the sharp dichotomy once posed between moments of openness when
2
institutions are crafted and highly constrained processes of reproduction once institutions are in
place. Instead, these arguments seek to show what kinds of changes propelled by what kinds of
actors or forces are most likely under what kinds of institutional configurations. An important
strand of this research, focused on some developing nations, documents and explains a surprising
lack of change in political outcomes despite significant formal-institutional change.1 Our work,
focused on contexts in which formal rules are in fact observed and enforced, grapples with the
opposite reality: frequently overlooked but very significant changes in political outcomes despite
formal-institutional stability.2 Crucially, these processes typically play out over long periods of
time, which makes them visible only to scholars who are sensitive to temporal variation over a
considerable duration. For this reason, comparative-historical analysis is uniquely suited to
capturing and explaining these processes, so often missed by other modes of analysis.
In the next three sections, we outline the scope and importance of drift and conversion,
the basic structural conditions under which we should expect each, and the similarities,
differences, and relationships between the two. In the penultimate section, we lay out some
common lessons. A crucial shared feature of drift and conversion is that they “turn the tables” by
creating a new status quo that reverses who is playing offense and defense. Protectors of the old
status quo, the last round’s winners, are now the losers. Combatting drift (updating the institution
to reflect the environment) or “correcting” conversion (realigning the institution with its prior
goals) requires assembling a new coalition, which is often extremely difficult.
1
The work of Levitsky and his colleagues, for example, has shown that in developing countries formal
(“parchment”) institutions that contradict deeply entrenched informal institutions (conventions, norms) often do not
produce the expected, or really any change at all in established political dynamics (Levitsky and Slater 2011: 1; see
2
Many institutionalists follow North in adopting a broad definition of institutions that includes both formal and
informal arrangements, but the work of Levitsky and Slater shows that we are far better off if we do not conflate the
two under the same heading. Only by distinguishing formal (obligatory) institutions from informal norms (based on
shared understandings and conventions) can we explore the connections and relationship between the two.
3
Because of this shared characteristic, drift and conversion elevate actors with specific
strategic imperatives and capacities. Blocking action to foster drift is generally much less
“traceable”—to use Douglas Arnold’s phraseology (Arnold 1990)—than creating new
institutions or policies, and so actors pursuing non-majoritarian aims may find it particularly
attractive. Conversion often takes place in venues that are harder to monitor (such as courts and
bureaucracies) and places high informational demands on participants. This makes it a
potentially quite effective means of changing an institution or policy in ways that are difficult for
citizens to recognize or defenders to reverse.
It also makes drift and conversion particularly attractive to organized “repeat players” in
political life. In American politics research—and, increasingly, comparative politics research as
well—dominant scholarly perspectives focus on shifts in voting and public opinion. For the
kinds of subterranean changes we examine, however, voters are generally not the prime movers,
though they may play an important conditioning role. Instead, we argue for a renewed attention
to the strategies and capacities of organized interests that are aware of long-term opportunities
inherent in existing formal rules. Thus, a research program that takes drift and conversion
seriously will necessarily return to fundamental questions of political economy, examining who
has power to shape outcomes over the long term due to their disproportionate organizational
resources or occupation of strategic nodes in the process of reform.
DRIFT, CONVERSION, AND INSTITUTIONAL CHANGE
Drift and conversion are difficult cases for institutional theory because they represent
instances in which stability in formal institutions belies significant changes in what an institution
4
does or what effects it has. By “formal institutions,” we mean formalized rules that are in
principle obligatory and subject to third-party enforcement. Without denying the importance of
informal conventions, the arguments and conjectures that we elaborate are subject to specific
scope conditions; they assume the rule of law and state capacity to enforce formal rules.
Furthermore, we believe it makes sense to think of some, but not all, public policies as
constituting formal institutions according to this definition, insofar as they embody legally
enforceable rules, create new organizations with state-backed decision or enforcement power, or
both (Pierson 2003). Thus, policies that set rules for social interaction that are enforced through
the exercise of public authority qualify under our definition, as do organizations or agencies (for
example, the American National Labor Relations Board) whose existence is publicly guaranteed
and whose actions are backed up by state authority. This conceptual expansion makes clear that
some policies and organizations constitute enduring features of the political landscape that
should be studied in similar fashion to the traditional state institutions on which institutional
work focuses. It also highlights the oft-neglected links between theories of legislative politics,
which focus on the production of authoritative public laws, and theories of institutional design,
which are typically concerned with more encompassing political institutions.
Although drift and conversion are frequently overlooked, the tools of institutional
analysis provide important grounding for understanding them. Both drift and conversion occur
under circumstances well explored by institutionalists—when, because of institutional and
partisan roadblocks and vested defenders of the status quo, authoritatively replacing existing
institutions or policies is off the table. They are, in short, strategies for changing outcomes
without altering formal rules. In the case of drift, the trigger for change is context discontinuity,
5
the occurrence of environmental shifts that existing institutions are poorly adapted to handle. In
the case of conversion, the trigger is actor discontinuity, as actors not involved in (in some cases
not even around for) those rules’ creation seek to redirect the institution to pursue ends distinct
from those it was originally set up to produce. We consider each process in turn.
Drift
Drift is defined as the failure of relevant decision-makers to update policy or institutional
rules to reflect changing social circumstances in ways that are recognized by at least some
political actors and consequential for the effects of those rules on society (Hacker 2004, 2005;
Hacker and Pierson 2010, 2011; for extensions, see Streeck and Thelen 2005; Lynch 2006;
Beland 2007; Mahoney and Thelen 2009; Falleti and Lynch 2009). By formal rules, we mean the
specific operating language created by authoritative public decisions. By changing social
circumstances, we mean the shifting context of those rules. Drift occurs when such shifts alter
policy or institutional outcomes without a change in formal policy or institutional rules (Hacker
2004, 2005; Streeck and Thelen 2005).
Drift is therefore more than just political inaction. It requires (1) that the circumstances
around policies or institutions change in ways that alter the effects of those policies or
institutions on the ground, (2) that this change in outcomes is recognized, (3) that there are
alternative rules that would reduce the degree to which these shifts in outcomes occur (in other
words, that the shifts are potentially remediable); and (4) that efforts to update these rules fail.3
3
Another common aspect of drift, which makes it both more interesting and more subversive of existing models,
that the failure to update formal rules is not explained by the preferences of democratic majorities but instead by the
blocking activities of intense minorities or actors possessing veto powers. In a prior formulation (Hacker and
6
A dramatic example is the difficulty that established welfare states have had grappling
with newly intensified economic risks (Esping-Andersen 1999). Where established programs do
not cover emerging new risks, the role the welfare state plays in citizens’ lives will shrink
“naturally” over time (Hacker 2004; Taylor-Gooby 2004). Conservative parties in Europe that
come to power on platforms promising to reduce government spending may not wish to engage
in costly political fights to dismantle popular social programs, but they can pursue their
objectives by simply declining to update these programs. In cases such as this, drift reflects the
different priorities of a new majority coalition, and is chosen as a strategy because it advances
this new agenda without provoking the same political fallout that would accompany more direct
retrenchment measures.
Yet drift also comes in a “non-majoritarian” variant. In 1997, the Chair of the U.S.
Federal Reserve, Alan Greenspan, observed a growing mismatch between a dynamic economy
and sclerotic policy-making—and applauded:
With technological change clearly accelerating, existing regulatory structures are being
bypassed, freeing market forces to enhance wealth creation and economic growth… As
we move into a new century, the market-stabilizing private regulatory forces should
gradually displace many cumbersome, increasingly ineffective government structures.
This is a likely outcome since governments, by their nature, cannot adjust sufficiently
quickly to a changing environment, which too often veers in unforeseen directions
(Quoted in Johnson and Kwak 2010, p. 101).
Greenspan’s faith in “market-stabilizing private regulatory forces” proved badly misplaced, of
course. Yet his pronouncement accurately captured the degree to which federal financial
policies—failing to adjust “sufficiently quickly to a changing environment”—were simply being
Pierson 2011), we made this last feature part of the definition of drift. We prefer the looser definition here, as it
recognizes that democratic majorities may support the inaction that causes drift. As we shall argue, however, drift is
a particularly potent way for political actors to pursue change that would likely not command majority support if
carried out in more overt ways.
7
swept aside. While deregulation played an important role in prompting these changes, at least as
important was the failure of policymakers to respond to new market behaviors and technologies,
in part because of intense lobbying by the financial industry to head off such responses. Instances
like this of non-majoritarian drift occur when well-placed minority interest are able to exploit
possibilities to block change even where a majority might prefer otherwise.
Conversion
By “conversion” we mean the transformation of an already-existing institution or policy
through its authoritative redirection, reinterpretation, or re-appropriation. Thus conversion, like
drift, combines elements of constancy in institutional form with change in institutional impact.
Drift, however, involves active resistance to the adaptation of formal rules to new circumstances.
Political actors seeking drift need only to block change. By contrast, conversion requires actively
shifting the purposes to which existing institutions or policies are put. The hallmark of
conversion is that it allows reformers to pursue important substantive changes even in the face of
formidable obstacles to more direct forms of institutional re-engineering.
Conversion reminds us that institutions are not usually dedicated resources; they are
“multipurpose tools” that can be used to different ends—and whose goals and functions are
therefore contested by groups both outside and within them (Perrow 1986: 11-12).4 Not only do
institutions have multiple possible effects, these effects are frequently unanticipated (especially
over long periods), in part because the actors charged with shaping and operating institutions are
often distinct from those who create them (Pierson 2004: esp. chapter 4). Conversion occurs
4
The reference in Perrow is to what he calls “complex organizations” but the insights apply much more generally as
we argue below.
8
when those actors are (1) able to redirect an institution or policy to serve ends distinct from those
of its creators while (2) leaving its formal rules in place.
Like drift, conversion comes in a majoritarian and a non-majoritarian variant. As an
example of the former, when the British Labour Party came to power in 1945, it inherited a range
of policy instruments that had been created as a result of wartime emergency powers. Rather
than dismantle these institutional capacities the party instead redirected them to serve new,
peacetime goals. Acting on its longstanding ambitions to use state power to redistribute wealth,
the party harnessed these policies to massively increase the role of the state in the economy and
to assign to it a wholly new role—stabilizing markets, regulating imbalances, administering
newly nationalized industries (Shonfield 1969; Judt 2008). In cases such as this, conversion
strategies are often a matter of political expediency. Their main advantage is to allow
institutional innovators to emphasize continuity so as to enlist the support of those who are
invested in the existing institutions (or at least to avoid their active opposition), while also
neutralizing other groups who might otherwise be alarmed at the prospect of change.
In other cases (such as the Sherman Antitrust Act), conversion strategies are pursued
more surreptitiously by political actors with sufficient organization, resources, and permanency
to alter or even subvert an institution’s original purposes or functions. The development of the
European Union provides a telling example. When the member states of the (then) EEC adopted
Article 234, this included a “preliminary ruling mechanism” that would allow them to use the
European Court of Justice to “hold EU legislative and executive bodies in check” (Alter 2000:
491). Clearly designed as a mechanism to allow member states to resist excessive
encroachments on the part of the supranational bodies of the EU, domestic actors within these
9
countries seized on this provision to challenge national policies they opposed. The successes of
these groups before the European Court of Justice transformed the preliminary ruling mechanism
from a tool for reaffirming member sovereignty into one for powerfully increasing the
obligations of the member states under EU law (Alter 2000; Abbott and Snidal 2000: 438).5 As
this example suggests, non-majoritarian variants of conversion often involve a shift in the arena
of contestation, away from the more visible venues (in this case, domestic legislatures) to less
visible arenas (the European Court) that reward groups who possess technical or informational
advantages.
When Do Drift and Conversion Occur?
We have already suggested that drift and conversion occur when the opportunities for
formal revision of policies or institutions are limited. We can state this more specifically by
saying that these are modes of change favored when the “status quo bias” of the political
environment is high. A large institutional and choice-theoretic literature has explored the
institutional arrangements, procedural rules, and distribution of political preferences that foster
such stasis in authoritative decision-making. Tsebelis’ work on veto players, for example,
associates heightened policy stability with contexts in which more actors must accede to change,
when the ideological distance between them is greater, and when they are more internally
cohesive (Tsebelis 2005; see also Immergut 1992). Thus status quo bias results from the
interaction of decision-making institutions and the distribution of preferences. The political
institutions of the United States, featuring bicameralism and frequent divided government, as
5
Conversion strategies of this sort need not be attempts at wholesale reversals, of course. They can also take the
form of changes that bend or stretch rules beyond their original intent.
10
well as such supermajority hurdles as the Senate filibuster and presidential veto, create a high
barrier to the enactment of new legislation. In recent decades, these institutional hurdles have
been greatly accentuated by rising partisan polarization (McCarty, Poole, and Rosenthal 2006).
Though the United States is an extreme case, as a rule democratic polities provide opportunities
for intense minorities to block change, even if that change has the actual or potential support of
popular majorities. In such cases, there is no guarantee that alternatives to the status quo that
would garner the support of popular or legislative majorities will be put in place.
Of course, decision-making rules and political polarization are just two crucial factors
affecting status quo bias. As the literature on institutional stability has shown, replacing
institutions or policies outright often involves significant costs and risks even when the political
landscape is more forgiving. Among these barriers are uncertainty about the effects that
institutional changes will have (Sheple 1986), the vested interests created by the feedback effects
of existing policies (Pierson 2000), the efforts of the original creators of the institutions or
policies to “hard-wire” them to reduce the chance they will be unraveled by present and future
opponents (Moe 1989), the crowding of agendas that makes it difficult to revisit past decisions
(Kingdon 1984; Baumgartner and Jones 1993), the collective action problems and transaction
costs of creating new coalitions, and the ever-present danger of causing “collateral damage”
when institutions or policies are tightly inter-coupled.
Thus, we see a crucial similarity between drift and conversion: they are fostered by
political settings that make authoritative change more difficult. The crucial difference between
drift conversion concerns the degree to which the institution or policy under challenge can be
altered without replacing it through the creation of new formal rules. Borrowing the language of
11
legal scholars, we call this second dimension the “precision” of an institution or policy.6
Whereas the first dimension calls attention to the political context of the formal rules that are the
object of possible reform, this second dimension draws attention to the character of the formal
rules themselves. Policies and institutions vary in the extent to which they afford actors
discretion in their interpretation or enforcement—whether, in the words of Kenneth Abbott and
his coauthors, the specify “clearly and unambiguously what is expected of a state or other actor
(in terms both of the intended objective and the means of achieving it) in a particular set of
circumstances” (Abbott et al. 2000: 412). The rule that stipulates that employers must pay a
wage of at least $7.25 per hour is precise; the rule that renders actions “in restraint of trade”
illegal much less so.
Policies whose provisions are ambiguous (and contested) and whose effects depend on
interpretation and implementation by other actors offer fertile terrain for strategies of conversion.
Such strategies thrive on rule ambiguity and they frequently take place in venues—courts and
bureaucracies—where these built-in ambiguities are worked out. By contrast, policies whose
rules are unambiguous are less susceptible to such strategies, though they are potentially more
vulnerable to drift, because what is not in the rule is also not covered by the rule. Treaties with
the U.S.S.R. about arms control, for example, were very precise. For this very reason, however,
they also left unregulated all sorts of behaviors that were outside the specific provisions of the
treaty.
Status quo bias and precision are dimensions, not dichotomies. They can best be thought
off as posing tradeoffs. Status quo bias protects policies from direct revision, but makes
6
In previous work we have called this “barriers to internal change.” Here we adopt the concept of precision as
defined by legal scholars such as Kenneth Abbott, Anne-Marie Slaughter, and Thomas Franck.
12
strategies of change through drift and conversion more attractive and more difficult to reverse
through authoritative updating or replacement of formal rules. Greater precision reduces
ambiguity but also the adaptability of rules to new circumstances.7 By contrast, less precise rules
are more flexible but also more susceptible to conversion, because they require someone to work
out what the rule calls for in a specific situation.
Where Drift and Conversion Meet: The Politics of Delegation
To the extent that existing scholarship addresses these tradeoffs, it does so under the
broad heading of “delegation.” An essential problem faced by policymakers is that it is difficult
ex ante to specify what particular policy steps need to be taken to achieve policy goals,
especially when the social context of policymaking is unpredictable and front-line decisions
require specialized information.8 In such circumstances, elected officials may wish to delegate
power to bureaucrats who can use their greater information to respond to changes in the
environment. Alternatively, or actually concomitantly, the rule can be framed in such a way that
it allows these front line agents the flexibility they need to adapt the rule to changing
circumstances.
A large body of scholarship has discussed the circumstances under which elected officials
delegate power to front-line policy administrators, frequently employing the familiar principalagent model to capture the process (for a good review, see Bendor, Glazer, and Hammond 2001).
7
Similarly, as a large literature on incomplete contracting tells us, actors often prefer precision to increase
enforceability, but such precision carries own problems—among others, rendering agreements more brittle and
increasing the possibility that bargaining partners will not be able to reach any agreement in the first place (Abbott
and Snidal 433ff).
8
There is also a vast literature on “incomplete contracting” that deals with these problems.
13
In fact, analysts employing the principal-agent framework have actually used the term “drift,”
though not as we do (for a recent important exception, see Callendar and Krehbiel 2012). They
have identified two sorts of drift of particular interest to them—bureaucratic drift and coalitional
drift (Shepsle 1992). Bureaucratic drift refers to the classic principal-agent problem: Bureaucrats
have their own interests; using their discretion, they may pursue goals different from those that
elected officials support. Coalitional drift, by contrast, taps into many of the same processes to
which we draw attention with the concept of “conversion.” It refers to the prospect that
dominant political coalitions will shift over time, meaning that the framers of a policy may not
be in control when further political interventions occur down the road. We take up these two
possibilities—bureaucrats pursuing their own interests, and new coalitions attempting to gain
control of previously created institutions or policies—in turn. To avoid confusion, we refer to the
former as “bureaucratic shirking: and the latter as “coalitional change,” reserving the phrase drift
for failure to adapt policies or institutions to their shifting environment in consequential ways.
The problem of bureaucrat shirking has been central to the literature on delegation. Yet
this body of scholarship is largely silent on how and why drift occurs in the face of
environmental change, devoting much of its energies instead to showing why drift is not really a
problem. Stressing the foresight and capacity of legislative principals, the delegation literature
suggests that politicians have a grab bag of means of aligning agencies’ interests with their own
to update a policy or institution to continue to pursue its original goals in a changing world—
means that often go beyond formal legislative language to include both formal and informal ex
post controls. For example, they can empower affected interest groups to shape agency decisions
on an ongoing basis. Or they can adopt recruitment or oversight processes that encourage
14
bureaucrats to align their goals with those of politicians. In essence, the delegation literature
implies that bureaucratic drift is generally preventable through ex ante and ex post actions. Even
without enacting new policies, legislators can get front-line policy actors to do what they want.
Yet this view is far too sanguine about the ability to forestall or respond to drift. The
problem goes beyond aligning the incentives of bureaucrats with those of their principals. Unless
the discretion granted to administrators is sufficient and appropriate to meet the particular
environmental shift confronted, it cannot solve the problem. In many of the cases we discuss
below, involving responses to major demographic or technological change, discretion would
have needed to be exceptionally wide and flexible. In these cases, discretion may only be
effective by transferring policymaking authority to bureaucrats, essentially positioning
themselves as unconstrained legislators. Yet this is exactly the threat that the literature on
delegation sought to eradicate. The dilemma of drift cannot easily be elided with discretion.
What about coalitional change? Here the problem is similar. While coalitional change is
seen as central to the principal-agent dilemma, scholars have been far too sanguine about the
capacity to forestall conversion by tying the hands of competing coalitions in the future. We
think this is primarily because the literature has a limited amount to say about the different sorts
of political actors who are active during implementation as opposed to enactment and the
strategies they employ to convert policies or institutions to new ends. The implication in much
of this work is that policy choices are majoritarian bargains that bake into the legislative cake
both policy intentions and a set of controls on implementers that keep future discretionary action
within the bounds of the enacting coalition’s original goals (Callendar and Krehbiel 2012). Yet
the political forces shaping legislative outcomes are generally far different than those shaping
15
implementation: public salience is much lower, issues much more technical, and organized
interests much more active and influential (Culpepper 2011). To be sure, members of the
enacting coalition are not unaware of this reality, but there may be real limits to their ability to
foresee or forestall future developments at odds with their original goals amid the far less
majoritarian politics of implementation, especially when these processes plays out over long
periods of time.
After all, it is not just bureaucrats who are charged with working out how rules should be
interpreted and applied to new situations. Where rules are formulated so as to grant front-line
administrators more flexibility to adapt them to local context, the resulting ambiguities
frequently encourage challenges from outside actors. In many countries, the courts play the
crucial function of providing the “authoritative” interpretation. In this way as well, delegation as
a “solution” to bureaucratic drift often simply moves the politics into other, less visible venues,
providing fertile soil for actors whose goals are not congruent with the legislation to promote
substantive outcomes that can be quite distant from the original intent of the law.
In short, the principal-agent framework provides an important catalogue of potential
remedies for bureaucratic shirking and coalitional shifts, but it provides limited purchase on the
overlapping politics of drift and conversion. In the next two sections, we seek to provide that
purchase, developing arguments about the circumstances under which these two modes of
change are most likely to emerge. These two sections underscore what is missed in institutional
analyses not attuned to these two forms of change and their interrelation and underscore the
benefits of an expanded institutionalist research program that takes them seriously.
16
DRIFT
Drift is a ubiquitous feature of advanced industrial democracies. As political polarization
increases and public programs mature, it has become harder and harder to update policies to
reflect the dramatic changes in economies and societies that have played out over the last
generation. This gap is particularly notable in the United States. Over the last two decades, a
large body of research has emerged in the study of the American Congress analyzing the forces
that create (or break) legislative “gridlock” (McCarty, Poole, and Rosenthal 2006; Wawro and
Schickler 2010). This research has highlighted in particular the growing polarization of
American legislators and the increasing use of filibuster in the Senate, which allows opponents
of all non-treaty legislation besides the annual budget to demand a 60 percent majority for
passage. Within this literature, gridlock is typically assumed to reinforce the status quo. Yet drift
suggests a large exception to this assumption: When the effects of policy or institutions are
heavily dependent on their environment, gridlock can produce major changes in the status quo,
as failure to adapt policies or institutions leads to substantial shifts in their outcomes. Doing
nothing in Congress means doing something in the world.
Indeed, one of the exciting features of drift as a theoretical concept is its potential for
linking fairly circumscribed arguments about legislatures and other decision-making institutions
to substantively interesting social phenomena by incorporating aspects of the broader social
context. Indeed, as Falleti and Lynch argue, “Drift is a mechanism that can operate only in a
system characterized by multiple layers of relevant context” (Falleti and Lynch 2009, p. 1157,
emphasis added). We do not just have to say “drift happens.” We can also say something about
the major kinds of social dynamics that are likely to give rise to this kind of mismatch. Here we
17
attempt only to sketch the broad types of environmental change and main features of policies that
are likely to engender drift.
Drift and the Environment
The effects of policies or institutions are often changed by environmental shifts that the
original designers of the policy or institution have difficulty fully anticipating or preparing for.
Though the sources of contextual change are multiple, three loom large in any consideration of
drift: technological change, demographic shifts, and differential growth rates.
Rapid technological change can often introduce new possibilities for social organization
or interaction that radically alter the functioning of existing policies. The development of the
network of technologies associated with “containerization,” for example, led to plummeting
prices for long-range shipping, producing an ever-widening ripple of social consequences, from
the proliferation of low-cost imports that posed huge competitive challenges to domestic
manufacturers in rich nations to the rise of low-cost forms of retailing based on these products
(Levinson 2006). In the United States, the technological revolution in shipping posed a radical
challenge to the existing regulatory structure, which was fundamentally ill-equipped for the new
realities. A similarly radical process has played out in the realm of finance. As Alan Greenspan
noted, technological development—specifically developments in information technology—
turned financial markets, and the meaning of financial regulations, upside down. The
reconfiguration of finance initially occurred largely outside the contours of regulatory policy and
institutions, although eventually this asymmetry led to growing pressure for statutory adjustment
18
(Johnson and Kwak 2010). Either way, these developments fundamentally changed the effects
(and effectiveness) of existing rules.
As with technology, so too with demography. Many policies are implicitly designed
around particular distributions of actors or activities within a population. Over time, however,
these distributions may undergo dramatic changes that radically transform the impact of those
policies. In many democratizing nations in the nineteenth and early twentieth centuries, property
and income restrictions for voting were not fixed in nominal terms and assumed a small pool of
wage laborers; as a result, the burgeoning working class created by industrialization could be
brought into the electorate without explicit expansion of the franchise. In the United States,
population shifts over many decades have vastly increased the extent to which the Senate over
represents sparsely populated states. At the time of the nation’s founding, the ratio of the mostpopulous to least-populous state was about 17-1. It is now about 70-1 (Lee and Oppenheimer
1999).
Differential growth rates are a final powerful source of drift. Many policies fail to cover
all activity in a particular issue area, or cover different aspects in different ways. Where these
differences in coverage are associated with differential growth rates, the result can be substantial
drift. Policies that operate in one fashion when the composition of actors or activities have a
particular distribution may operate very differently when that distribution changes. In the United
States, for example, industrial relations policies that were geared towards manufacturing and
allowed considerable state-level variation underwent great change with no alteration of the
governing statutes as population and employment drifted towards states and sectors with low
rates of union coverage and greater obstacles to organizing. Once this shift in distribution
19
reached a tipping point, it radically destabilized the industrial relations system – even in the
absence of formal changes in policy. Similarly, in the field of private pensions, traditional rules
grew gradually outmoded as more and more employers shifted from heavily regulated definedbenefit plans offering a fixed benefit in retirement to defined-contribution retirement accounts
(for example, 401ks) (Hacker 2004).
A shared and distinctive feature of all these contextual shifts is that they involve “big,
slow-moving processes” (Pierson 2004) that most analyses ignore. The study of drift thus can
contribute to a less abstract and formal political science that systematically situates political
struggles within evolving social contexts and, through that connection, addresses substantive
political conflicts (Hacker and Pierson 2009). Thinking about drift allows us to form clear
expectations about when and why such contextual changes matter. It can also help us think more
clearly about our next topic, the relationship between drift and policy design.
Drift and Design
The external environment matters enormously in creating the conditions for drift. Yet
elements of design also play a critical role. Perhaps the most important of these is
automaticity—the degree to which policies are hard-wired to change in response to specified
environmental shifts. 9
9
We will put to the side one feature of policy that can obviously lead to a shift in the policy status quo without new
legislation—namely, explicit limits on the duration of policies. A law may create new benefits for only ten years, or
require reauthorization after five. In such cases, policymakers need to pass new legislation if they wish to continue a
policy. Such “temporal bounding,” however, is not equivalent to drift. Drift occurs when policy outcomes change
due to environmental shifts. The process may be gradual and not immediately perceptible. A pure time limit is more
like an on-off switch; one day the policy is in place, the next it is not. Moreover, such temporal limits are explicitly
written into law, and thus relatively easy to see and mobilize around. The same is generally not true of drift, which
20
As R. Kent Weaver argues (1986), one of the purest cases of automaticity is indexation—
the explicit tying of policies to quantifiable changes in environmental conditions. The most
common variant of this is the incorporation of automatic inflation adjustments into key policy
elements, such as benefit levels, eligibility criteria, and relevant income brackets. Given how
common indexation is today, it is a surprisingly recent development. Before its widespread
adoption, lawmakers would have to periodically increase benefits if they were to remain more or
less constant in real dollars. With indexation, the default became an inflation-tied increase, and
any departure from that default required legislative action.
Indexation is perhaps the most well-known of a class of automaticity provisions that
update a policy on the basis of indicators relevant to its performance, such as wage growth,
inflation, and demographic change. Another common provision in the American context is
entitlement status, which simply means that everyone eligible for a benefit under its terms
automatically receives it, without additional enactments.10
Automatic provisions all pose the same basic tradeoff: In return for giving up some
measure of control over policy, legislators free themselves from the need to enact periodic
updates that require overcoming the hurdles to new legislation. For this reason, automaticity
generally reduces the chance of drift. When policies are hard-wired to reflect shifts in the
external environment, there is much less likely to be a departure from the policy’s original aims
due to contextual changes.
results from the over-time interaction of environmental change and often complex features of policies—including
features that are not written into law, such as the absence of automatic inflation updates.
10
There is also a legal meaning of entitlement, related to the ability to make claims in U.S. courts—we focus on the
policy meaning here.
21
Automacity is not a panacea. In many policy areas, as we have seen, it will be difficult or
unappealing for elected officials to specify ex ante what particular steps would need to be taken
to prevent drift. In such instances, policymakers may wish to write into the legislation some
adaptive flexibility, by framing the rules themselves in terms of general principles rather than
detailed provisions, by delegating power to front-line agents, or both. But while delegation is
sometimes seen as a solution to drift, it elevates the alternative venues and attentive political
actors that we have already highlighted, thus opening the door to conversion strategies as
organized groups attempt to exploit their superior expertise and access to influence to change
institutions or policies from within. In this sense, delegation may simply promote other forms of
hidden change, notably through strategies of conversion.
CONVERSION
In many respects, conversion is the flip-side of drift. Drift is a strategy of holding firm
while the world changes. Conversion is a strategy that taps opportunities for altering what
institutions and policies do without rewriting the rules themselves. If the possibility of drift
arises when contexts change, opportunities for conversion arise when actors change—when
political actors inherit institutions and are bound by rules not of their own making and
inconsistent with their ends. Conversion allows reformers to sidestep many of the collective
action problems and political costs associated with efforts to replace institutions outright.
We have already discussed the considerable barriers to dismantling existing institutions
and rules. Yet if dismantling institutions is difficult, constructing new ones is even more
challenging. Even when vested interests can be conquered, the prospect of institutional
22
innovation can galvanize previously neutral groups who begin to contemplate the ways in which
their interests might be compromised by new institutional arrangements. This applies not just to
the target of institutional reform itself, but also to any other adjacent policies or arrangements
that might experience some form of collateral damage. Previously neutral parties often emerge in
defense of existing institutions or policies even if the impact of the proposed change on their own
interests is indirect or unknown (Mahoney and Thelen 2012, chapter 1). For all these reasons,
reformers may have strong incentives to avoid frontal assaults and seek conversion instead.
What makes conversion more or less possible? Here again, we highlight two main
factors: rule ambiguity and the opportunities that a polity provides for opponents of existing
policies or institutions to find alternative venues in which to pursue hidden change.
Conversion and Rule Ambiguity
Conversion as a political strategy begins with the observation that institutional rules are
sometimes imprecise and thus allow for significant leeway in their interpretation and
enforcement. Especially in democratic contexts, policies and legislative rules are frequently
messy compromises, containing inconsistencies and contradictions. Sometimes, for example,
decision makers will intentionally leave aspects of a policy vague to make it easier for competing
interests to “sign off” on a measure, or to enhance the durability of the rule by leaving room for it
to be extended and applied to new situations. Whatever the source, ambiguity provides a
strategic opening for interested parties to push substantive outcomes in their preferred direction,
because inevitably political actors have different preferences over how the rule should be
understood and applied (its “meaning”—that is, what it requires and rules out). Questions of
23
interpretation in these cases are inherently political, since actors with divergent interests will
seek to invoke their preferred interpretation, and indeed impose it on others.
That such opportunities are rarely lost on political actors is illustrated by the increased
use in the United States of presidential “signing statements.” First employed by James Monroe
in the early nineteenth century, signing statements historically had played a largely ceremonial or
rhetorical function, allowing presidents to assign or claim credit for particular pieces of
legislation as he formally signed them into law (Kelley and Marshall 2009: 172). Beginning
under Ronald Reagan, however, these statements began to develop “into a significant and
commonly used instrument for executive direct action” (Cooper 2005: 517). By adding
“clarificatory” language regarding how the law should be enforced, presidents could actively
shape the implementation of legislation and thereby alter its substantive impact. Used
aggressively, signing statements allow the president to steer outcomes toward a new policy status
quo (Kelley and Marshall 2009) and one that can be fiendishly difficult for opponents to defeat
precisely because it happens after the fact and mostly out of the public glare.11
The observation that conversion is a preferred strategy for situations or for actors who
prefer to effect change beneath the radar leads to our next topic, the role of venue-shopping.
11
George W. Bush was famous for using his veto power very sparingly. However, he signed a record-breaking
number of statements that in effect made the enforcement and implementation of certain aspects of the affected laws
subject to presidential discretion. The strategic use of presidential signing powers allowed Bush to avoid the use of
the more visible (and possibly divisive) presidential veto. Vetoes are blunt instruments (presidents have to veto the
whole law, whereas signing statements allow for more selective interventions). They are often unpopular and
controversial. But perhaps most important, using signing statements instead of the veto denies Congress the
opportunity to vote to override.
24
Conversion and Venue-Shopping
Perhaps the most important features of conversion as a concept is that it draws our
attention to arenas of political contestation that remain largely out of view in most of the
mainstream literature. The heavy emphasis on highly visible political theater within legislatures
and parliaments over major reform initiatives obscures the massively consequential pulling and
hauling that ensues after a law is passed. As the work of Aaron Wildavsky, Shep Melnick and
Eric Patashnik (among others) have all reminded us, “The passage of a reform law is only the
beginning of a political struggle” (Patashnik 2008: 3).12 Where legislative “re-dos” are not
possible, reformers may shift their attention to other arenas, among which bureaucracies and the
courts stand out as particularly important to the politics of conversion.
Conversion as a political strategy generally involves shifting substantive issues into
particular arenas (or keeping them out of particular arenas) so as to marginalize or exclude
potential opponents while also finding the venue that best “fits” one’s own particular resource
endowments. Bureaucratic agencies, for example, are often a crucial arena of contestation and
conversion. Judith Layzer’s Open for Business (2012) for example explains how conservative
interests have advanced policy change in the area of environmental regulation through “lowprofile policy challenges” at the agency level. “Although conservatives have failed to repeal or
revamp any of the nation’s environmental statutes,” she writes, “they have influenced the
implementation of those laws in way that increased the risks we face, prevented or delayed
12
This point is echoed, from an international relations perspective, by Joensson and Tallberg, who note that
“agreements – whether in political or legal garb, whether on national or international levels—do not put an end to
political processes” (1998: 373).
25
action on newly recognized problems, and altered the way Americans think about environmental
problems and their solutions.”
Alongside bureaucratic agencies, the courts are a second crucial arena of contention—and
conversion—in many countries. Courts, after all, are in the business of interpreting what a law
or policy calls for or requires, and judicial rulings on such matters often have an enormous
impact on substantive outcomes. A large literature on political jurisprudence has provided crucial
insights into how statutory interpretation and judicial influence have shaped important areas of
public policy in the United States. In the area of civil rights, court decisions in the United States
famously pushed public policy beyond where legislative majorities were prepared to go. The
1964 Civil Rights Act—a deeply controversial piece of legislation that outlawed various forms
of discrimination—was hotly contested and very nearly failed in the Congress. The
compromises that had been required to get the law passed had weakened many of its key
provisions (Lieberman 2007; Frymer 2003). Once enacted into law, however, civil rights
lawyers proceeded to aggressively probe the bounds of the legislation by bringing a steady
stream of cases before the courts. The result was to establish a “robust private enforcement
regime” that transformed an initially weak legislative rule into one of the strongest affirmative
action regimes in the world (Frymer 2007: 84).
Conversion is not just a strategy for weak interests. It is available and attractive to the full
range of well-organized groups, and those with substantial resources and lobbying and legal
capacities—particularly employers and their associations—often loom largest in the politics of
conversion. For example, in a recent study of the politics of the Federal Communications
Commission, John de Figueiredo and Emerson Tiller show that firms have been able to change
26
the regulatory environment in which they operate through the strategic uses of litigation to
“overturn potentially harmful regulations passed by federal administrative agencies” (2000: A1).
In another recent study, Lee Epstein and colleagues find that business interests in the United
States have been unusually successful under the Roberts court in securing favorable
interpretations of legislation of importance to them (2012: 1471 and passim).
In short, as a theoretical concept conversion forces us to look beyond the high-profile
politics of legislatures and parliaments that dominate the headlines and the mainstream literature.
It directs our attention to these less visible—but highly consequential – arenas where the politics
continue long after the enactment of the formal laws. The balkanization of political science,
featuring separate literatures on “public administration” and “the courts,” obscures the political
dynamics that come into view once we see bureaucracies and courts not as arenas apart from and
above the political fray but instead as deeply imbricated in a network of political institutions that
jointly shape substantive outcomes (Carpenter 2001; Kagan 2003; Melnick 1994; Barnes 2007;
Burke and Scherer 2012).
THE POLITICS OF HIDDEN CHANGE
By drawing attention to what we are calling the politics of hidden change, drift and
conversion have several important implications for the study of politics. As just discussed, they
help make clear the crucial significance of political venues beyond legislatures. In doing so, they
also shine a bright light on the otherwise murky and underappreciated role of organized interests
in institutional and policy change. Finally, recognizing the possibilities for hidden change shifts
27
our view of who is playing “offense” and “defense” in democratic politics. Conversion and drift
allow the political tables to be turned in ways that are frequently difficult for those on the losing
side of these shifts to recognize or reverse. When opponents of existing policies or institutions
are able to block updating or successfully pursue strategies of conversion—or even to shape the
social environment around policies or institutions directly so that they drift from their original
purposes—supporters of the status quo ante (the winners in the last round) are faced with
difficult challenge of remobilizing to return policies or institutions to their prior state. In politics,
there is a huge difference between playing offense and defense. Drift and conversion change who
is in the driver’s seat, magnifying the power of actors who are best able to bring them about.
Looking Beyond Big Legislative Changes
When studying institutional and policy change, scholars mostly focus on episodes of
explicit policy reform when formal rules are rewritten. Such studies have shed much light on
how institutional arrangements facilitate or impede major policy shifts (as in the “veto points”
literature (Immergut 1992)) and how coalitional realignments make possible “breakthrough”
legislation or allow for more gradual policy shifts even in areas that were previously seen as
impervious to change (e.g., Häusermann 2010).
By contrast, drift and conversion draw our attention to the politics beyond what Hacker
and Pierson (2010) call “electoral spectacle” and the usually highly visible politics within
legislatures and parliaments. They are powerful reminders that the politics are by no means over
when the gavel comes down. As Patashnik emphasizes, “The losers from reform cannot be
counted on to vanish without another fight, and new actors may arrive on the scene who seek to
28
undo a reform to further their own agendas” (2008: 3). Well-organized and well-resourced
interest groups that lack a large social base may in fact wish to avoid big legislative battles,
which can be self-defeating if they stir up media and popular attention or mobilize citizens or
other groups with competing interests. Where this can be anticipated, it may make very good
sense to avoid a frontal assault on existing rules and seek change instead through “quieter”
political channels (Culpepper 2011).
Thus, even beyond the possibility of legislative rollback or erosion, institutional and
legislative reforms are potentially subject to subtle but significant change as the venue shifts— to
the bureaucracies charged with implementing legislative rules and to the agencies (including
courts) charged with interpreting them. Moving to these non-majoritarian venues alters the
political dynamics substantially, not least by taking these issues even further from public view.
Conversion and drift require capacities and reflect strategies that are often different from those
involved in winning big legislative fights, especially when those fights are salient. Political
actors seeking drift or conversion need to have the ability to monitor policies and institutions
across multiple venues over the long term, and the organizational prowess to prevent the
updating of existing rules or to shape the way such rules are enacted on the ground. They also
generally have an interest in promoting change that occurs quietly and without prominence—the
fait accompli rather than the legislative landmark.
The Role of Organized Interests
Who are the political actors most likely to possess these attributes? The short answer is
organized interests. They, as Terry Moe (1989, 269) puts it, “are normally the only source of
29
political pressure when structural issues are stake.” On issues that are low profile, where the
balance of organized groups is highly skewed, and where the link between legislative action (or
inaction) and outcomes remains complex and opaque, it is organized interests rather than voters
whose influence is likely to loom largest.
Of course, these are precisely the main attributes of drift and conversion that we have
charted. Evolving often slowly, recognized only episodically, if at all, by voters, and devilishly
difficult to trace back to particular political leaders, drift and conversion are the terrain where the
advantages of organization for long-term monitoring, foresight, and coordination are likely to be
most pronounced. On this terrain, politicians seeking to be responsive both to organized groups
and to atomized voters will find the sway of the former much stronger than the pull of the latter.
Because it is so hard to link the changes produced by drift back to particular elected officials, this
form of change is perhaps the least risky way for politicians in democratic polities to cater to
organized interests.13 Similarly, conversion strategies that operate through influence over the
enforcement or interpretation of specific rules or policies are often lost completely on voters with
their notoriously short attention spans and limited appetite and aptitude for technical and legal
detail.
Thus, given the discussion so far, organized groups are likely to loom large. But which
groups and why? To guide our thinking on such matters we can draw on insights drawn from a
large literature on policy reform and public law. In different ways Patashnik ( 2008—from a
policy reform vantage) and Stuart Scheingold (2004—from a public law perspective) have both
13
The legislative politics literature ignores this conditional nature of group influence in part because it assumes that
both groups and voters factor into legislators' ideal points in a more or less stable fashion. But as a result, this work
misses how legislators can shift their focus (groups vs. voters) and thus their positions relatively quickly in those
rare cases when voters gain the upper hand vis-à-vis groups.
30
noted that even big victories (for Patashnik, major legislative reforms; for Scheingold landmark
court cases) may be deeply satisfying in the moment, but are never in themselves sufficient to
alter political outcomes. Both emphasize the possibility of considerable slippage after the
legislation is passed or the ruling is handed down. Scheingold emphasizes the important role of
social mobilization as a necessary complement to strategies based on legal activism (see also
Barnes and Burke 2006: 497; also Epp 1998; 2009). Scheingold finds that without high levels of
social mobilization to ensure consistent and vigorous implementation and enforcement, even big
legal victories may prove pyrrhic. Patashnik draws the same conclusion, arguing that “even the
most solid reform ideas require ongoing collective support to endure” (Patashnik 2008, xi).
Thus, arguing from different vantage points, Patashnik and Scheingold converge on the
conclusion that reforms that are premised on strong social mobilization are less fragile than those
based on weak political coalitions premised on expediency.
Moreover, these analyses also stress that concentrated groups enjoy distinct advantages
over diffuse interests in the politics after legislative or judicial breakthroughs (e.g., Patashnik
2008, 33). A classic article by Marc Galanter (1974) makes this point within the public law
literature. Galanter draws a broad distinction between what he calls “one shotters” and “repeat
players,” arguing that repeat players enjoy myriad advantages over one shotters. More
important, he claims that these advantages cumulate over time, therefore resulting in a gradual
shift of the law in favor of the repeat players. Among the advantages repeat players enjoy are:
Superior expertise: Repeat players enjoy economies of scale in terms of developing inhouse legal expertise. Moreover, because they have numerous cases under their belts, repeat
players are in a better position both to anticipate and head off problems (for example, avoiding
lawsuits through particular contractual language) and to prevail in the cases that do arise. With
31
this experience in turn come many opportunities to develop relationships with institutional
incumbents and build a reputation that then becomes an asset in the game itself.
Long time horizons: Repeat players can almost by definition play for the long term. This
involves a cluster of strategies and accompanying advantages. They can play the odds and aim
to maximize gains over a series of cases rather than go for a big victory on a single case.
Crucially, they can play for rules instead of playing for immediate gains. In other words, they
can settle --in poker terms, “fold”—when victory seems out of reach and the danger exists that a
loss will result in unwelcome precedents, or conversely, accept a loss if it comes bundled with a
more favorable interpretation of the rule of which they have run afoul.14
One shotters are disadvantaged on both counts. First, by definition, they lack equivalent
experience. Even if they happen to command a superior grasp of the substantive issue at stake,
they lack equivalent legal experience (and associated reputational effects). Second, one shotters
have shorter time horizons and thus more limited strategic options. Actors who are engaged in a
one-off contest obviously cannot play the odds strategically across a number of cases. Moreover,
in one-shot contests the cost of defeat is likely large given the sunk investment in a single
particular battle. Just as poker amateurs can have a good night but over the long run will lose to
professionals, one shotters may win particular cases but over time the cumulative results should
favor repeat players.
We think that these insights extend beyond the realm of the courts and carry broad
lessons for strategies of conversion and drift, and for political outcomes more broadly. Elections
are blunt instruments and voters are classic one shotters whose influence is felt only sporadically
and in ways that are far removed from actual policy outcomes. Organized interests by contrast
are repeat players par excellence. Their presence is virtually permanent, regardless of the
political stripe of the government, their influence is targeted at specific legislative and policy
14
An example is a Supreme Court decision on sexual harassment in which the case was won by the victim but in
which the majority opinion (written by Justice Scalia) makes harassment very hard to demonstrate in subsequent
rounds.
32
outcomes, and they are in it for the long run and well positioned to follow legislation into the
implementation and enforcement stages. One shotters backed up by strong social mobilization,
like the amateur poker player mentioned above, may be able to score a victory in some particular
case, but they are unlikely to have ongoing durable success unless they develop the kind of
experience that could turn them into repeat players (Barnes and Burke 2006: 497).
Re-Conceiving Defense and Offense
Both drift and conversion often involve a switch in which groups are on offense and
defense. Perhaps the clearest finding of institutional scholarship is that there are very large
differences between attacking and defending in politics. Drift and conversion are strategies for
policy change that capitalize on this inherent asymmetry. In the language of nondecision
theorists, they reflect the “mobilization of status quo bias.” When policies are vulnerable to drift,
the coalition that was previously seen as playing offense (attacking the status quo) may in fact be
playing defense (protecting a drifting policy from revision). Once the minimum wage has begun
to erode though drift, for example, workers will progressively lose ground unless their political
allies remobilize to change the law. Similarly for conversion, once a new interpretation is in
place—through agency action or court decision—this effectively shifts the policy status quo, and
opponents of the original rule will have put themselves in the (easier) defensive position for the
next round. To return to the example with which we began, in the case of the Sherman Antitrust
Act, it took nearly a quarter of a century before a new law (the Clayton Act) was finally passed
as a modest correction that protected unions from some of the specific types of judicial
injunctions that had been visited upon them in the wake of the Sherman Act.
33
Probing even more deeply, many organized interests can exert substantial influence on
the social environment as well as on politics. Large employers and unions, for example, are not
simply collective players in the political realm; they are also capable of coordinated action in the
economic realm (Lindblom 1982; Offe and Wiesenthal 1980). For these actors, it may be
possible to achieve policy goals not by changing law but by “breaking out” from it: changing
their behavior while preventing the updating of law. In the late 1970s and early 1980s, for
example, American employers adopted a more aggressive stance toward unions in the private
sector, then backed up their private actions with a strong political push to resist union’s efforts to
update industrial relations law. Thinking systematically about these forms of private power, and
how they play into policymaking, could provide an important means of giving specific content to
arguments about the “structural” influence of political actors that once featured prominently in
political and social analysis (Hacker and Pierson 2002).
CONCLUSION AND IMPLICATIONS
Until recently, institutionalism was dominated by the analysis of “comparative statics.”
Institutions were associated with stability and invoked as an independent or intervening variable
to explain persistent cross national differences in some other (policy or political) outcome. More
recent work has attempted to move beyond this traditional comfort zone of comparative statics to
gain more insight into how institutions themselves evolve and change over time. The concepts
of drift and conversion nicely connect these old and new lines of theorizing by focusing on one
important variety of bounded innovation—strategies that are pursued in contexts in which more
thoroughgoing institutional overhaul is politically too costly or difficult. As concepts, they allow
34
us to navigate between structural determinism and unbridled agency by helping us to pinpoint
what specific moves are possible under what kinds of conditions.
Drift and conversion also provide a crucial link between the analysis of institutional
change and that of institutional reproduction, two issues that are too often treated as separate.
Particularly in punctuated equilibrium models of change, scholars are prone to separate the
analysis of innovation (where agency often looms large) from the analysis of subsequent
institutional reproduction (where the emphasis is on structural constraints on action). In fact,
however, reproduction and change are two sides of the same coin (for a more extended
argument, see Thelen 1999; 2004): Understanding the particular mechanisms of reproduction on
which an institution rests will provide insight into the specific factors or conditions that could
provide an opening for change. Conversely, institutions do not survive long stretches of time by
standing still: As the world around them changes, their survival rests not on faithful
reproduction, but on ongoing active adaptation to their political and economic environment.
Thus while our emphasis has been on drift and conversion as modes of change, both
concepts also provide crucial insight into institutional reproduction. Drift signals the hollowing
out of a reform though inaction, reminding us that reproduction is not a simple matter of stasis
but requires active interventions to adapt these institutions and rules to changes in the political
context. Conversion is often a strategy through which this is accomplished, as new groups adapt
institutions inherited from the past to new purposes and in so doing contribute to the survival of
institutions and rules not of their own making (Thelen 2004).
Finally, drift and conversion expand our range of vision by prompting us to adjust not
just what kind of change we are looking for, but where we are looking for it and whom we
35
expect to produce it. As we have seen, both draw our attention to forms of politics that
mainstream research in comparative politics and especially American politics frequently
overlook. For example, by problematizing issues of interpretation and enforcement, conversion
draws our attention to oft-neglected political arenas—especially the courts and bureaucracies—
that powerfully shape substantive outcomes. In this way, the concept offers the opportunity to
tap into well-developed traditions of research on bureaucracy and jurisprudence while inviting a
more sustained dialogue with scholars in public law, public administration, and public policy
whose work sometimes seems disconnected from the rest of political science.
Above all, drift and conversion bring us back to fundamental question of political
economy. In a world marked by stark information and power asymmetries, by vast differences in
resources and organizational capacities, who actually governs (Dahl 1961)? Political contestation
often pits the organized and information-rich—the political actors most able to block the
updating of policies, the actors most capable of changing institutions from within—against the
disorganized and information-poor (Arnold 1990; Bawn et al 2012; Cohen et al 2008; Hacker
and Pierson, 2005; 2010). By focusing our eyes on previously neglected forms of change, where
they occur, and who typically brings them about, we can better understand not just the sources of
institutional change but also the structure of power in advanced societies.
36
REFERENCES
Abbott, KennethW., Robert O. Keohane, Andrew Moravcsik, Anne-Marie Slaughter,
and Duncan Snidal, 2000. The Concept of Legalization. International Organization 54(3): 17-35.
Arnold, R. Douglas. 1990. The Logic of Congressional Action. New Haven: Yale University
Press.
Bachrach, Peter, and Morton S. Baratz. 1962. “Two Faces of Power.” American Political Science
Review 56(4): 947-952.
Barnes, Jeb. 2007. "Bringing the Courts Back In: Interbranch Perspectives on the Role of Courts
in American Politics and Policy Making." Annual Review of Political Science 10:25-43.
Barnes, Jeb, and Thomas Burke. 2006. "The Diffusion of Rights: From Rights on the Books to
Organizational Rights Practices." Law and Society Review 40 (3):493-524.
Beland, Daniel. 2007. “Ideas and Institutional Change in Social Security: Conversion, Layering,
and Policy Drift.” Social Science Quarterly 88 (1): 20-38.
Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas and Instability in American Politics.
Chicago: University of Chicago Press.
Bawn, Kathleen, et al. 2012. “A Theory of Political Parties: Groups, Policy Demands and
Nominations in American Politics.” Perspectives on Politics 10(3): 571-97.
Bender, Jonathan, A. Glazer, T. Hammond. 2001. “Theories of Delegation.” Annual Review of
Political Science 4: 235-269.
Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent: Logical Foundations
of Constitutional Democracy. Ann Arbor: University of Michigan Press.
Binder, Sarah. 2003. Stalemate: Causes and Consequences of Legislative Gridlock. Washington,
DC: Brookings.
Brady, David, and Craig Volden. 1997. Revolving Gridlock. Boulder, CO: Westview Press.
Burke, Thomas, and Nancy Scherer. 2012. "The Bush Administration and the Uses of Judicial
Politics." In Building Coalitions: Making Policy: The Politics of the Clinton, Bush and
Obama Presidencies, ed. M. Levin, D. DiSalvo and M. Shapiro. Baltimore: Johns
Hopkins University Press.
37
Callander, Steven, and Keith Krehbiel. 2012. "Gridlock and Delegation in a Changing World."
Stanford School of Business Research Paper No. 2100, March.
Carpenter, Daniel. 2001. The Forging of Bureaucratic Autonomy: Networks, Reputations and
Policy Innovation in Executive Agencies. Princeton, NJ: Princeton University Press.
Cohen, Marty, David Karol, Hans Noel, and John Zaller.. 2008. The Party Decides. Chicago:
University of Chicago Press.
Cooper, Phillip J. 2005. "George W. Bush, Edgar Allan Poe, and the Use and Abuse of
Presidential Signing Statements." Presidential Studies Quarterly 35 (3):515-32.
Culpepper, Pepper. 2011. Quiet Politics and Business Power. New York: Cambridge University
Press.
Dahl, Robert. 1961. Who Governs: Democracy and Power in an American City. New Haven:
Yale University Press.
de Figueiredo, John M.. 2000. Litigating Regulation: Corporate Strategy in
Telecommunications. Academy of Management Proceedings,
http://web.mit.edu/jdefig/www/papers/litigation_regulation.pdf, A1-A6,
DiMaggio, P., and Walter W. Powell. 1991. "Introduction." In The New Institutionalism in
Organizational Analysis, ed. W. W. Powell and P. DiMaggio. Chicago: University of
Chicago Press.
Elster, Jon. 1979. Ulysses and the Sirens. New York: Cambridge University Press.
Epp, Charles. 1998. The Rights Revolution: Lawyers, Activists and Supreme Courts in
Comparative Perspective. Chicago: University of Chicago Press.
———. 2009. Making Rights Real: Activists, Bureaucrats and the Creation of the Legalistic
State. Chicago: University of Chicago Press.
Falleti, Tulia, and Julia Lynch. 2009. “Context and Causal Mechanisms in Political Analysis.”
Comparative Political Studies 42(9):1143-1166.
Frymer, Paul. 2003. "Acting When Elected Officials Won't: Federal Courts and Civil Rights
Enforcement in US Labor Unions, 1935-85." American Political Science Review 97:48399.
Frymer, Paul. 2007. Black and Blue: African Americans, the Labor Movement and the Decline
of the Democratic Party. Princeton: Princeton University Press.
38
Galanter, Marc. 1974. "Why the "Haves" Come out Ahead: Speculations on the Limits of Legal
Change." Law and Society Review 9 (1).
General Accounting Office (GAO). 2006. Mandatory Spending: Using Budget Triggers to
Constrain Growth. Available at www.gao.gov/new.items/d06276.pdf.
Gibson, Edward. 2012. Boundary Control: Subnational Authoritarianism in Federal
Democracies. New York: Cambridge University Press.
Hacker, Jacob S. 2001. “Learning from Defeat? Political Analysis and the Failure of Health Care
Reform in the United States,” British Journal of Political Science 31(1): 61-94.
________. 2002. The Divided Welfare State. New York: Cambridge University Press.
________. 2004. “Privatizing Risk without Privatizing the Welfare State: The Hidden Politics of
Social Policy Retrenchment in the United States.” American Political Science Review
98(2): 243-260.
Hacker, Jacob. 2005. "Policy Drift: The Hidden Politics of US Welfare State Retrenchment." In
Beyond Continuity: Institutional Change in Advanced Political Economies, ed. W.
Streeck and K. Thelen. Oxford: Oxford University Press.
Hacker, Jacob S., and Paul Pierson. 2002. “Business Power and Social Policy: Employers and
the Formation of the American Welfare State.” Politics & Society 30(2): 277-325.
________. 2005. “Abandoning the Middle: The Bush Tax Cuts and the Limits of Democratic
Control.” Perspectives on Politics 3(1): 33-55.
________. 2009. “The Case for Policy-Centered Political Analysis,” Paper Presented at the
APSA Annual Meeting, Toronto.
________. 2010. Winner-Take-All Politics. New York: Simon and Schuster.
________. 2011. “Drift and Democracy: The Neglected Politics of Policy Inaction,” Paper
Prepared for the Working Group on Institutional Change, Cambridge, MA, February.
Hamilton, Alexander. 1787. Federalist. 23. Available at www.constitution.org/fed/federa23.htm.
Hardin, Russell. 1989. “Why a Constitution?” In Bernard Grofman and Donald Wittman, eds.,
The Federalist Papers and the New Institutionalism. New York: Agathon.
39
Häusermann, Silja. 2010. The Politics of Welfare State Reform in Continental Europe. New
York: Cambridge University Press.
Heclo, Hugh. 1974. Modern Social Politics in Britain and Sweden: From Relief to Income
Maintenance. New Haven, CT: Yale University Press.
Helmke, Gretchen, and Steven Levitsky, eds. 2006. Informal Institutions and Democracy:
Lessons from Latin America. Baltimore: Johns Hopkins University Press.
Huntington, Samuel P. 1971. "The Change to Change: Modernization, Development, and
Politics." Comparative Politics 3(3): 283-322
Immergut, Ellen. 1992. Health Politics: Interests and Institutions in Western Europe. New York:
Cambridge University Press.
Johnson, Simon, and James Kwak. 2010. 13 Bankers: The Wall Street Takeover and the Next
Financial Meltdown. New York: Pantheon.
Jacobs, Alan. 2009. “Policymaking as Political Constraint: Institutional Development in the U.S.
Social Security Program,” in Explaining Institutional Change: Ambiguity, Agency and
Power, eds. James Mahoney and Kathleen Thelen, 94-131. New York: Cambridge
University Press.
Judt, Tony. 2008. Postwar: A History of Europe Since 1945. New York: Penguin.
Kagan, Robert A. 2003. Adversarial Legalism and American Government. Cambridge, MA:
Harvard Univeristy Press.
Karol, David. 2001. How and Why Parties Change Positions on Issues: Party Policy Change as
Coalition Management in American Politics. Presented at Annual Meeting of APSA in
San Francisco.
Kelley, Christopher S., and Bryan W. Marshall. 2009. "Going it Alone: The Politics of Signing
Statements from Reagan to Bush II." Social Science Quarterly 91 (1):168-87.
Kingdon, John, W. 1984. Agendas, Alternatives, and Public Policies. New York: Harper Collins.
Knight, Jack. 1992. Institutions and Social Conflict. New York: Cambridge University Press.
Krehbiel, Keith. 1998. Pivotal Politics: A Theory of U.S. Lawmaking. Chicago: University of
Chicago Press.
40
Lee, Frances E., and Bruce I. Oppenheimer. 1999. Sizing up the Senate. Chicago: University of
Chicago Press.
Levine, Ross. 2010. “An Autopsy of the US Financial System: Accident, Suicide, or Negligent
Homicide?” National Bureau of Economic Research Working Paper No. 15956, available
at www.nber.org/papers/w15956.pdf.
Levinson, Marc. 2006. The Box: How the Shipping Container Made the World Smaller and the
World Economy Bigger. Princeton: Princeton University Press.
Levitsky, Steven, and Maria Victoria Murillo. 2009. "Variation in Institutional STrength."
Annual Review of Political Science 12.
Levitsky, Steven, and Dan Slater. 2011. "Ruling Poltiics: The Formal and Informal Foundations
of Institutional Reform." In Ruling Politics, ed. S. Levitsky and D. Slater. Harvard
University.
Lieberman, Robert C. 2007. "Private Power and American Bureaucracy: The EEOC and Civil
Rights Enforcement." Columbia University.
Lindblom, Charles E. 1982. “The Market as Prison.” Journal of Politics 44 (2): 324-336.
Lynch, Julia. 2006. Age in the Welfare State: The Origins of Social Spending on Pensioners,
Workers, and Children (Cambridge: Cambridge University Press).
Madison, James. 1788. Federalist 58. Available at www.constitution.org/fed/federa58.htm.
Mahoney, James, and Kathleen Thelen, eds. (2009). Explaining Institutional Change: Ambiguity,
Agency and Power. Cambridge: Cambridge University Press.
McCarty, Nolan. 2007. “The Policy Consequences of Political Polarization.” In The
Transformation of the American Polity, eds. Paul Pierson and Theda Skocpol. Princeton:
Princeton University Press.
McCarty, Nolan, Keith T. Poole, and Howard Rosenthal. 2006. Polarized America. Cambridge:
MIT Press.
Melnick, R. Shep. 1994. Between the Lines: Interpreting Welfare Rights. Washington, DC:
Brookings.
Moe, Terry M. 1989. “The Politics of Bureaucratic Structure,” in Can the Government Govern,
eds. J. Chubb, P. Peterson. Washington, DC: Brookings Institution.
41
www.stanfordtechnologies.com/group/polisci/downloads/moeThe%20Politics%20of%20Bureaucratic%20Structure.pdf.
Moe, Terry. 2005. "Power and Political Institutions." Perspectives on Politics 3 (2):215-33.
North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance. New
York: Cambridge University Press.
Offe, Claus, and Helmut Wiesenthal. 1980. “Two Logics of Collective Action.” Available at
http://www.hwiesenthal.de/publik/hw/2logics.pdf.
Orren, Karen, and Stephen Skowronek. 1994. “Beyond the Iconography of Order: Notes for a
“New Institutionalism,” in The Dynamics of American Politics, eds. L. Dodd and C.
Jillson, 311- 330. Boulder, CO: Westview Press.
Patashnik, Eric M. 2008. Reforms at Risk: What Happens After Major Policy Changes are
Enacted. Princeton, N.J.: Princeton University Press.
Perrow, Charles. 1986. Complex Organizations: A Critical Essay. Third Edition ed. New York:
McGraw-Hill.
Pierson, Paul. 2000. “Path Dependence, Increasing Returns and Political Science.” American
Political Science Review 94 (2): 251-267.
________. 2001. “The Problem of Democratic Control in an Age of Big Government,” in
Politics at the Turn of the Century, eds., A. M. Melzer, J. Weinberger, and M. R. Zinman.
Lanham, MD: Rowman and Littlefield.
________. 2003. “Public Policies as Institutions. ” Paper Prepared for the Yale Conference on
Crafting and Operating Institutions, New Haven, April.
________. 2004a. Politics in Time: History, Institutions, and Social Analysis.Princeton:
Princeton University Press.
Rae, Douglas W. 1975. “The Limits of Consensual Decision.” American Political Science
Review 69 (4): 1270-1294.
Rehder, Britta. 2006. "Recht und Politik beim Wandel des Flächentarifs. Juristen als politische
Akteure im System der Arbeitsbeziehungen " Politische Vierteljahresschrift 47 (2):16992.
Schattschneider, E. E. 1960. The Semi-Sovereign People. New York: Holt, Reinhart, and
Winston.
42
Scheingold, Stuart. 2004. The Politics of Rights: Lawyers, Public Policy, and Political Change.
Ann Arbor: University of Michigan Press.
Schickler, Eric. 2001. Disjointed Pluralism. Princeton: Princeton University Press.
Schumpeter, Joseph. 1942. Capitalism, Socialism, and Democracy. New York: Harpers.
Shepsle, Kenneth A. 1986. "Institutional Equilibrium and Equilibrium Institutions." In Political
Science: The Science of Politics, ed. H. Weisberg. New York: Agathon.
Shierholz, Heidi. 2009. "Fix It and Forget It: Index the Minimum Wage to Growth in Average
Wages." In EPI Briefing Paper, ed. E. P. Institute. Washington, DC: Economic Policy
Institute.
Shonfield, Andrew. 1969. Modern Capitalism: The Changing Balance of Public and Private
Power. London: Oxford University Press.
Steinmo, Sven, Kathleen Thelen, and Frank Longstreth, eds. 1992. Structuring Politics:
Historical Institutionalism in Comparative Politics. New York: Cambridge University
Press.
Streeck, Wolfgang. 2012. "A Note on Evolution." In Instituitonal Change Working Group.
Cologne, Germany.
Thelen, Kathleen. 1999. "Historical Institutionalism in Comparative Politics." The Annual
Review of Political Science 2.
Thelen, Kathleen. 2004. How Institutions Evolve: The Political Economy of Skills in
Comparative-Historical Perspective. New York: Cambridge University Press.
Stimson, James A., Michael B. MacKuen, and Robert S. Erikson. 1995. “Dynamic
Representation.”American Political Science Review 89:543-565.
Streeck, Wolfgang, and Kathleen Thelen. 2005. Beyond Cotinuity: Institutional Change in
Advanced Political Economies. Oxford: Oxford University Press.
Taylor-Gooby. Peter. 2004. “New Social Risks in Post Industrial Society.” International Social
Security Review 57 (3): 45-64.
Teles, Steven M. 2010. The Rise of the Conservative Legal Establishment: The Battle for Control
of the Law. Princeton: Princeton University Press.
43
Thelen, Kathleen. 2004. How Institutions Evolve: The Political Economy of Skills in Germany,
Britain, the United States, and Japan. Cambridge: Cambridge University Press.
Theriault, Sean. 2005. The Power of the People. Columbus, OH: Ohio University Press.
________. 2008. Party Polarization in Congress. Cambridge: Cambridge University Press.
Tsebelis, George. 1995. Decision Making in Political Systems: Veto Players in Presidentialism,
Parliamentarism, Multicameralism and Multipartyism. British Journal of Political Science
25 (3): 289-325.
44