When the Smoke Clears: Expertise, Learning, and Policy Diffusion Charles R. Shipan * University of Michigan Craig Volden The Ohio State University October 29, 2010 Abstract In federal systems, governments have the opportunity to learn from the policy experiments – and the potential successes – of other governments. Whether they seize such opportunities, however, may depend on the expertise or past experiences of policymakers. Based on an analysis of a dataset on state-level adoptions of youth access antismoking adoptions, we find that states are more likely to emulate other states that have demonstrated the ability to successfully limit youth smoking. In addition, we find that political expertise (as captured by legislative professionalism) and policy expertise (as captured by previous youth access policy experiments at the local level) enhance the likelihood of emulating policy successes found in other states. * An earlier draft of this paper was presented at the 2007 Annual Meeting of the American Political Science Association, Chicago. The authors thank the Robert Wood Johnson Foundation for financial support, Jamie Chriqui for providing us with the updated version of the National Cancer Institute’s State Cancer Legislative Database, and Fred Boehmke, Rob Franzese, George Krause, Paula Lantz, and seminar participants at the University of Chicago, the University of Michigan, the University of Pittsburgh, Stanford University, the University of Texas, and the University of Virginia for helpful comments. In addition, local tobacco control ordinance data was provided by the American Nonsmokers’ Rights Foundation Local Tobacco Control Ordinance Database © .
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
When the Smoke Clears: Expertise, Learning, and Policy Diffusion
Charles R. Shipan* University of Michigan
Craig Volden
The Ohio State University
October 29, 2010
Abstract
In federal systems, governments have the opportunity to learn from the policy experiments – and the potential successes – of other governments. Whether they seize such opportunities, however, may depend on the expertise or past experiences of policymakers. Based on an analysis of a dataset on state-level adoptions of youth access antismoking adoptions, we find that states are more likely to emulate other states that have demonstrated the ability to successfully limit youth smoking. In addition, we find that political expertise (as captured by legislative professionalism) and policy expertise (as captured by previous youth access policy experiments at the local level) enhance the likelihood of emulating policy successes found in other states.
* An earlier draft of this paper was presented at the 2007 Annual Meeting of the American Political Science Association, Chicago. The authors thank the Robert Wood Johnson Foundation for financial support, Jamie Chriqui for providing us with the updated version of the National Cancer Institute’s State Cancer Legislative Database, and Fred Boehmke, Rob Franzese, George Krause, Paula Lantz, and seminar participants at the University of Chicago, the University of Michigan, the University of Pittsburgh, Stanford University, the University of Texas, and the University of Virginia for helpful comments. In addition, local tobacco control ordinance data was provided by the American Nonsmokers’ Rights Foundation Local Tobacco Control Ordinance Database©.
When the Smoke Clears: Expertise, Learning, and Policy Diffusion
In federal systems, subnational governments have the potential to operate as policy
laboratories, experimenting with new ideas, abandoning failures, and exporting successes to
other governments. A key component of the spread or diffusion of successful policies from one
subnational government to the next is the ability of policymakers to learn from others’
experiments and to adapt policies to meet different circumstances at home. When legislators in
one state observe that another state has had success dealing with a policy problem, they may
choose to enact policies similar to those found in the successful state.
States may vary, however, in the extent to which they draw upon the policy experiences
of other states, with features of a state’s institutional and policy environment influencing the
extent to which it learns from the actions of others. As with confronting any problem where
learning is involved, the experience and expertise of policymakers may affect whether and how
they learn about policy successes elsewhere. To begin with, state legislators differ in their
political expertise, based, among other considerations, on whether lawmakers meet year round,
hold other jobs as well as being legislators, or have substantial staffs to aid in the acquisition of
information and expertise. In addition, states vary considerably in their policy expertise
regarding the particular policy in question. Notably, given the multiple levels of government in
American federalism, expertise may be gained through experience at the local level, with
localities conducting their own policy experiments.
One view of expertise is that it facilitates learning – experts can rely on their knowledge
and past experiences to adapt policies found elsewhere to circumstances at home. Another view
is that expertise may serve as a substitute for learning, with experts believing there is little left to
be learned from others beyond what they already know from previous experiences. In either
2
case, if expertise affects legislators’ abilities or willingness to learn from policy experiments that
have taken place elsewhere, then we can conclude that the diffusion of successful public policies
is not universal, but rather is a conditional process, with some governments choosing policies
that have been proven effective and others not learning from such successes.
In order to explore how learning affects the adoption of public policies, and how
expertise affects the learning process, we utilize a comprehensive dataset of laws from the 1990s
and early 2000s that aim to limit youth access to cigarettes. Our analysis contributes to the
understanding of policy diffusion in two ways. First, we examine whether states learn from
other states that have demonstrated policy success. In addition to showing that states are
influenced by interstate learning generally, we also demonstrate that our findings are robust to
alternative measures of success. Second, and more importantly, we show that the influence of
learning is indeed conditional, with both the political expertise and the policy expertise within
each state modifying the degree to which states emulate successful policies found elsewhere.
Diffusion, Learning, and Success
The literature on policy diffusion is vast and expanding rapidly. Fortunately, there are
now a number of useful and recent literature reviews on this topic (e.g., Berry and Berry 1999;
Karch 2007b in the American politics literature; Weyland 2005 and Meseguer and Gilardi 2009
in comparative politics; Simmons, Dobbin, and Garrett 2005 in international relations; and
Graham, Shipan, and Volden 2008 across subfields). Especially to the extent that they cover the
literature on diffusion across American states, these reviews identify seminal studies in this area
(e.g., Crain 1966; Walker 1969; Gray 1973), important methodological innovations (e.g., Berry
3
and Berry 1990; Berry and Baybeck 2005; Volden 2006), and conceptual advances (e.g., Braun
and Gilardi 2006; Shipan and Volden 2008).
Although our study draws on numerous insights from these earlier analyses, it moves
beyond most of them in its focus on the role played by success and in its investigation of the
ways in which learning from the success of others is conditional. Previous studies have not
ignored the concept of learning – indeed, thorough consideration and discussion of what it means
for one government to learn from another, and how this process occurs, can be found throughout
the literature (e.g., Meseguer 2005; Grossback, Nicholson-Crotty, and Peterson 2004; Weyland
2006; Mossberger 1999; Boehmke and Witmer 2004; Berry and Baybeck 2005; Shipan and
Volden 2006, 2008). But only a handful of studies (e.g., Volden 2006; Meseguer 2006; Gilardi,
Füglister, and Luyet 2009; Gilardi 2010), have conducted systematic, large-N analyses of the
effect of success on diffusion. And although one study – Gilardi’s (2010) recent analysis of
unemployment benefits in OECD countries – has explicitly focused on the conditional nature of
learning, none have considered the relationship between learning and expertise.
Instead, scholars generally have taken one of three approaches to deal with the role that
learning plays in diffusion. First, some studies simply have asserted that diffusion takes place
due to some combination of mechanisms – usually learning, economic spillovers, and imitation –
but have not attempted to disentangle these mechanisms (e.g., Berry and Berry 1990; Shipan and
Volden 2006; Karch 2007a). Second, other studies have attempted to measure learning
indirectly, either by utilizing a proxy measure designed to identify the situations in which
learning is likely to take place (e.g., Boehmke and Witmer 2004; Shipan and Volden 2008) or by
trying to pin down, as much as possible, other mechanisms, such as economic spillovers, and
then treating learning as the residual category (e.g., Berry and Baybeck 2005). Third, studies
4
have relied on extensive case studies, with the corresponding strengths and weaknesses of this
approach (e.g., Weyland 2006). On the one hand, case studies provide rich, nuanced analyses of
how learning can occur and what sort of effect successful policies have on adoptions by other
governments; but on the other hand, the measures of success may be subject to interpretation and
the generalizability of the conclusions may be open to question.
Given that most scholarship on diffusion has either explicitly or implicitly identified
learning as a central component of diffusion, why have so few studies attempted either to pin
down the exact relationship between success and diffusion or to isolate the conditions under
which states are likely to learn from the successes of other states? One problem is that in order
to conduct a large-N study of whether the success of policies affects their diffusion, three
conditions must be met. First, there needs to be a generally agreed upon and objective measure
of what constitutes policy success. Although in some areas success may be easy to define, in
others it ends up being a more nebulous concept. For example, what constitutes success when a
state adopts a lottery – the number of people who play and win, or the amount of money the
lottery brings in to the state’s coffers, or the effect on the overall state economy? Second, the
information about the measures of success needs to be publicly available, in order to facilitate the
ability of one state to learn from another. If a state can easily find out whether another state has
had success in dealing with some policy area, it is much more likely to use that information in its
own decision making. Third, it must be plausible that the adoption of some set of policies will
help the state to achieve success.
These conditions must be met in order to analyze the effects of successful policies of one
polity on the likelihood of adoption in another, and also to examine whether these effects are
conditional on the political and policy expertise of policymakers. In order to determine whether
5
learning is conditional on policy expertise or experience, two additional features must be met.
First, in the U.S. context, cities and states both must have jurisdiction to adopt laws about the
policy in question. Second, data on both internal local policies within the state and external
policies in other states must be available to decision makers and researchers alike.
Although these considerations have limited the amount of scholarship that assesses
learning-based policy diffusion, these conditions do not substantially diminish the breadth of
real-world examples of the phenomena we are interested in studying. Environmental, education,
crime, and labor policies, as well as numerous others, are adopted at multiple levels of
government in the U.S. and are ripe settings for policy learning.
Hypotheses
We start with the classic learning concept, which holds that a state can learn from other
states. Although policymakers can learn any of a number of things from other states, here we
focus on one central feature of learning: whether policies found in other states are associated
with success. If a state observes such success elsewhere, then it engages in policy learning
(Gilardi 2010) and is more likely to adopt the policies found in these successful states.
Evidence of policy learning arises when a successful government’s policies spread more
quickly and more completely to other governments than do less successful policies (e.g., Volden,
Ting, Carpenter 2008). Consider first the easiest hypothetical case, one with a single outcome
and a single policy: State A is trying to decide whether or not to adopt a specific policy, X, in
order to improve some policy condition, Y. If State A observes that State B has a desirable value
of Y, and knows that State B has adopted X, this will increase the odds that State A will also
adopt X. State A has, in effect, learned from State B’s success with X, and based on this learning
6
has decided to adopt the policy found in that other state. On the other hand, learning can cut in
the opposite direction: states may be less likely to adopt a policy found in an unsuccessful state.
Most public policies, however, are not so simple. More often there is a set of policies – a
policy profile – that in combination determine overall effectiveness. The logic of success-based
learning, however, is the same. If State A observes a particularly effective outcome in State B,
and further sees that this other state has adopted a set of policies that are likely to have
contributed to this better outcome, then it will adopt a similar policy profile, mimicking much, if
not all, of State B’s policy profile. Our first hypothesis spells out this relationship.
Learning Hypothesis: States are more likely to emulate other states that have demonstrated policy success.
The Modifying Effect of Political Expertise
The Learning Hypothesis posits a very direct, simple relationship, where one government
learns about policy from another successful government. Not all policymakers are the same,
however. One key aspect on which they vary is the expertise they have when making decisions.
In this paper, we explore two forms of expertise: political and policy, each taken in order.
Political expertise may quite reasonably vary substantially across state legislatures. For
example, some states pay their legislators well and provide them with large numbers of staffers
who can help them handle the duties of their jobs; other states pay little, or even nothing, and
provide little help in the way of staff. Some states have well-developed committee systems,
while others do not. And some state legislatures meet year round, while others meet only a few
weeks a year, or meet only every other year.
These differences across state legislatures are manifestations of legislative
professionalism, where legislatures that meet more frequently, pay higher salaries, and have
7
better support structures in terms of staff and committees are characterized as more professional
than other legislatures. Not surprisingly, scholars have found that differences in legislative
professionalism affect a range of political phenomena, with higher levels of professionalism
associated with more policy innovation (Kousser 2004), a higher likelihood of reelection (Berry,
Berkman, and Schneiderman 2000), the incidence of divided government (Fiorina 1994), and the
ability to place constraints on bureaucratic actions (Huber, Shipan, and Pfahler 2001).
More generally, as Squire has observed, professionalism is a form of political expertise
that affects a legislature’s “capacity to generate and evaluate information in the policymaking
process” (2008, 223). The exact effect of political expertise on the ability to learn from the
success of other states is, however, uncertain. On one hand, such expertise could increase the
likelihood that a state would follow the lead of other, more successful, states. To begin with,
state legislatures that are more professional will be more likely to identify which other states
have had policy successes and what policies those states have passed. In addition, the staffers
and committees in highly professional legislatures will be better able to evaluate the policy
experiments found in other states and to determine whether those successful experiments will
work in their own state. Finally, these highly professionalized legislatures will be more likely
have the time and other resources needed to pass new laws that are based on the information they
have learned about other states’ policy experiments.
In this first account, then, political expertise, as measured by legislative professionalism,
should increase the likelihood of learning from success, largely because – as Squire notes in his
definition – more professional legislatures have higher evaluative capacities. Squire also notes,
however, that highly professionalized legislatures are more likely to be able to generate their
own information. They do not need to look elsewhere for evidence of success, because they can
8
produce their own ideas about which policies are likely to work in their own states. In the area
of antismoking laws, for example, staffers can read the relevant medical and public health
literatures; committees can invite experts to provide testimony about the potential effects of
various policy options; and the legislature will have time to consider different alternatives. All
of this can be done without reference to what has happened in other states. Rather, it is the less
professionalized legislatures, which lack these abilities, that are precisely the ones that benefit
from the experiments conducted in other states and that are therefore more likely to learn from
successes in these other states. In this view, then, the effect of success should decrease with
political expertise (i.e., the effect of success is greater for less professionalized legislatures than
for more professionalized legislatures). We formulate these two views as competing hypotheses:
Political Expertise Hypotheses: (Enhanced Effect) Political expertise will enhance the likelihood of adopting policies found in successful states. (Diminished Effect) Political expertise will diminish the likelihood of adopting policies found in successful states.
The Modifying Effect of Policy Expertise
Policymakers vary not only in the political expertise found within their legislatures; they
vary also in the policy expertise they have attained through prior experience with each policy
area. Most notably, in federal systems, even prior to an upper-level government experimenting
with policies, lower-level governments may have adopted policies in a variety of forms. As
such, state policymakers can draw on experiences with policy adoptions within the state by
focusing on policy adoptions at the local (e.g., city or county) level.
As with political expertise, the modifying effects of policy expertise could cut in either
direction. First, there may be a substitution effect, where information obtained from internal
experience – laws passed at the local level – can substitute for that obtained from other
9
governments. After all, one of the benefits of internal learning is that the population that is
affected by local laws does not just look similar, and behave in a somewhat similar fashion, to
the state population (as might be the case when imitating another state) – instead, the affected
population actually is part of the soon-to-be-affected state population. Thus, policymakers with
expertise gained from observing internal experimentation with a policy may have less need to
rely on external sources of information. They already have firsthand knowledge of both the
policy and the political effects of adopting laws. Consequently, states with higher levels of policy
expertise may be less likely to emulate the approaches taken by other states that have
demonstrated policy success.1 In this view, then, the influence of learning from other states will
diminish as states gain more internal experience with a policy.
Second, there may be a complementary effect, where the effect of external success
increases as state lawmakers gain more policy expertise. Knowledge about policies, including
which policies have succeeded at the local level and why, provides them with a foundation to
search for the best form of such laws, which can be found in successful states. Greater internal
experience may lead to greater reliance on external success because states need to have their own
foundation of policy expertise in order to know what will, or will not, work in their state; this in
turn lets them know which other states to emulate. We sum up these two competing views:
Policy Expertise Hypotheses: (Substitution Effect) States with a greater degree of policy expertise are less likely to adopt policies found in successful states. (Complementary Effect) States with a greater degree of policy expertise are more likely to adopt policies found in successful states.
1 Consistent with Shipan and Volden (2006), we acknowledge that local adoptions could either stimulate statewide adoptions or take the pressure to act off of state policymakers. However, we are asking a different question: upon deciding to adopt a new policy, will states with substantial internally based policy expertise be more or less likely to learn from successes found elsewhere?
10
Youth Access Laws
To examine these hypotheses, we draw upon a set of antismoking restrictions known as
youth access laws. These laws tend to have two related goals. First, most of these laws are
geared toward making it more difficult for children to purchase cigarettes. Along these lines, for
example, states have enacted minimum age requirements for the purchase of cigarettes,
restrictions on the location of vending machines, and penalties for establishments that sell
cigarettes to those under the age requirements. Second, states also have passed laws that more
directly attempt to reduce youth smoking rates. Most notably, many states have laws that create
education and awareness programs aimed at teenagers and other children. Overall, then, youth
access laws are designed to both directly and indirectly reduce youth smoking rates.
The effects of these laws can be measured in two ways. First, with the passage of the
Synar Amendment in 1992, Congress required all states to monitor and reduce youth access rates
over time. Under this program, which is administered by the Substance Abuse and Mental
Health Services Administration (SAMHSA), a division of Health and Human Services, states
must actively demonstrate progress in reducing these rates, in order to qualify for federal funds
under the Substance Abuse Prevention and Treatment (SAPT) block grant program.2 In effect,
states must conduct sting operations (or other similar approaches), organized according to
specific protocols, to determine the rate at which teens are able to purchase cigarettes.
Beginning in 1996 or 1997 (depending on the state), all were required to report the results of
these sting operations to SAMHSA. Since the SAPT funds from the federal government have
totaled between $1.5 and $2 billion per year, states have a strong financial incentive to
participate, and all states have reported the findings of their investigations on a yearly basis.
2 For information about this program see http://prevention.samhsa.gov/tobacco/default.aspx.
11
Second, since 1991 the Centers for Disease Control and Prevention (CDC) has completed
and maintained the Youth Risk Behavior Surveillance System (YRBSS). As part of this system,
the CDC conducts the Youth Risk Behavior Survey on a variety of teenagers’ activities,
including smoking.3 Furthermore, the data from this survey is amassed at the state level. The
YRBSS data, therefore, provide state-level information on smoking rates among teenagers.
These measures – compliance with the Synar Amendment and the YRBSS data on
smoking rates – give us two separate, but related, indicators of state policy success. Reducing
the rates at which teens are able to purchase cigarettes and lowering smoking rates among teens
are clearly the goals of these programs; hence, we have clear, objective, and agreed-upon
measures of success, meeting the first requirement for an analysis of the diffusion of successful
policies. The second requirement is also clearly met: this information is easily available to
states, with the federal government posting the data online to facilitate easy access.
The third requirement is that there needs to be a clear link between enactment of these
policies and success – in other words, that youth access laws actually make it more difficult for
youths to purchase cigarettes.4 To begin with, laws that require people purchasing cigarettes to
show identification, that assess fines on stores that are caught selling cigarettes to minors, and
that restrict the placement of vending machines to establishments that minors do not frequent
(e.g., bars) are all plausibly likely to reduce youth access. In addition, the evidence suggests that
these sorts of laws can change the behavior of retailers, making it less likely that they will sell
cigarettes to people who are younger than the legal age (e.g., Forster and Wolfson 1998).
3 The specific survey question that we use asks teenagers whether they have smoked a cigarette within the past thirty days. The data can be accessed at: http://apps.nccd.cdc.gov/yrbss. 4 State and local governments share jurisdiction, allowing us to test our second hypothesis. In addition, concerns about economic spillovers across state lines play little role in the diffusion of youth access laws, which helps to narrow the focus to learning as a central mechanism.
12
Although few studies have examined the link between youth access laws and the rates of
illegal purchases by minors, public health scholars have conducted numerous studies of the
relationship between youth access laws and smoking rates among teens. Many of these studies
have found strong evidence that tobacco control policies aimed at teenagers do reduce rates of
smoking in this age group (e.g., Ross and Chaloupka 2004; Luke et al. 2000; but see also Lantz
et al. 2000) and that lowering the rate of sales to minors also lowers the prevalence of youth
smoking (e.g., Dent and Biglan 2004). Given the plausible case that youth access laws are
effective (and indeed that some types of restrictions may work better than others), given the
ability of states to discern which laws worked the best elsewhere, and given data about policy
adoptions at both the state and local levels, this policy area provides an ideal setting in which to
test hypotheses about learning and its conditional nature based on expertise.
Methods and Data
As we discussed earlier, there are very few large-N studies of the role that success can
play in the diffusion of public policies. Here we draw on a major exception, Volden’s (2006)
pioneering study of the Children’s Health Insurance Program. Building on Berry and Berry’s
(1990) introduction of event history analysis as a way to study diffusion, and drawing also on
studies from international relations that examine dyadic relationships between countries, Volden
constructs a series of state-level yearly dyads. Thus, each year of the dataset contains
observations corresponding to each state (State A) paired with each other state (State B).5 This
approach thus allows for the possibility that each state can learn from every other state in the
5 In 2000, for example, there are forty-nine observations for potential learning by Alabama – one observation in which Alabama is paired with Alaska, one in which it is paired with Arizona, one with Arkansas, and so on. Each of the fifty states is in a similar position. Overall, then, there are up to 50 × 49 state-level observations for each year.
13
political system.6 Hence, we can examine not only whether a state learns from its geographic
neighbors, but also whether it learns from those states that are similar in other ways, or that have
exhibited particular success in the area of youth access laws.
We follow Volden’s technique, albeit with two alterations. First, Boehmke (2009)
demonstrated that under certain conditions, the dyadic approach can produce estimates that lead
to spurious evidence of policy diffusion. More specifically, he shows that failure to control for
the opportunity to emulate another state’s policy can lead to misleading estimates. Fortunately,
he also identifies a straightforward remedy for this potential problem: conditioning on the
opportunity to emulate. In our analysis, therefore, we eliminate all observations in which State A
does not have the opportunity to adopt State B’s policies, which occurs when State A has already
adopted all of the youth access provisions found in State B. Second, because our focus is on the
extent to which adopters learn from other states under varying conditions, we limit our
observations solely to the cases where State A has made some policy change in the current year.7
Thus our analysis reveals not which states adopt any policy change, but rather, conditional on a
state making a policy change, whether that state learns about policy successes elsewhere and
whether such learning is enhanced or diminished based on political and policy expertise.
Dependent Variable
To develop a measure of state-level youth access laws, we began by identifying sixteen
different youth access laws that state legislatures have adopted, ranging from restrictions on the
6 Gilardi and Fuglister (2008) study health insurance subsidy policies across Swiss cantons with this method. Meseguer (2006) uses an alternative and complementary approach. 7 This limitation has the added benefit of removing from the dataset all years in which states did not hold legislative sessions or in which they held sessions that focused on specific topics only (which is harder to control for). If we focus only on states that met in regular session, rather than limiting our analysis to states that have taken any action, we obtain substantively similar results.
14
locations of vending machines to delivery restrictions to requirements that all young purchasers
show identification as proof of age.8 For our dependent variable, we need to measure the extent
to which each state moves toward each other state’s policy profile – that is, toward the set of
policies that the other state has previously adopted – in any given year. To determine the
complete list of youth access policies that each state has adopted, as well as the year in which
they were adopted, we relied on the State Cancer Legislative Database (SCLD), which is a
compilation of all state-level antismoking laws.9 We collected this information for all states over
the entire period of our study, which runs from either 1992 or 1997 (depending on which
measure of success we use) through 2002, the time period of the most dramatic change in state
youth access policies, and the period that contains the best available policy success information.
Next, in each of the sixteen categories of laws, we determine whether State A adopts a
policy change in a given year and, if so, whether that change moves State A toward State B or
away from State B. We then sum the number of categories in which State A moves toward State
B and also the number of categories in which it moves away from State B.10 For example,
consider a dyad that consists of Missouri and Iowa in 2001. In that year, Missouri passed a
comprehensive youth access law that included seven new policy components of the sixteen we
8 The sixteen aspects of youth access policy that we examine are: age requirements, youth penalties, free distribution restrictions, vending machine restrictions, out-of-package sales restrictions, ID requirements, sign posting requirements, vendor licensing requirements, vendor penalties, location restrictions, education and awareness activities, behind-the-counter sales requirements, delivery and shipping restrictions, task force authorization, random inspections, and bidi restrictions. Also clearly relevant to whether youths purchase tobacco is cost, which governments can affect through taxes. However, because tobacco taxes are adopted for a variety of reasons, extending beyond youth access considerations, we do not include taxes among our policy changes. Future work exploring the conditional nature of learning regarding tax policy (whether pertaining to youth access to tobacco or other areas) would be welcome. 9 SCLD data can be accessed at http://www.scld-nci.net/. 10 This approach captures the idea that the adoption by State A of a policy that State B has already adopted moves A toward B, while the adoption by A of a policy that B does not have may move A away from B.
15
include. Compared to the policies that Iowa had at the start of 2001, five of Missouri’s adoptions
moved it toward policies in Iowa (youth penalties, vending machine restrictions, location
restrictions, behind-the-counter-sales requirements, and random inspections) and two moved it
away (out-of-package sales restrictions and delivery/shipping restrictions).
To begin with, we create a simple, dichotomous measure that accounts for whether State
A in the dyad has moved toward State B on more policy dimensions than it has moved away
from State B. In our example above, the dependent variable, Movement Toward State B, takes a
value of 1 for the 2001 Missouri-Iowa dyad. More generally, whenever the movement toward
State B exceeds the movement away from State B, this dependent variable takes on a value of 1;
otherwise the dependent variable takes a value of 0.11
We also code a second dependent variable using this same information. However,
instead of treating the dependent variable as a dichotomous measure, where we simply identify
whether the majority of movement is toward or away from the second state, for this measure we
look at the amount of movement. In the example given above, for example, where Missouri
moves toward Iowa in five categories and away from it in two, this variable, Amount of
Movement Toward State B, takes a value of three (five minus two). This more nuanced measure
thus captures not only the direction of movement, but also the extent to which State A follows
(or retreats from) State B.12
Independent Variables: Measures of Success and Other State B Characteristics
11 Substituting an alternative dependent variable that captures whether State A moved toward State B in any of the policy categories yields substantively similar results. 12 Substituting an alternative dependent variable capturing solely the number of categories on which State A moved toward State B, rather than subtracting off movements away from State B, yields results that are similar substantively (although somewhat weaker statistically).
16
Because of the nature of our hypotheses, our primary independent variables are measures
of success across the states – namely, their success in reducing youth access and smoking. In
this section we discuss our measures of success, as well as two additional characteristics of State
B that increase the probability that State A will emulate its policy profile. In the following
sections we then discuss further control variables.
Of the variables characterizing conditions in State B, those most crucial to testing our
hypotheses are measures of success. The first of our success variables incorporates information
about whether states have prevented minors from purchasing cigarettes, which is the goal that the
Synar Amendment imposed on the states. We construct this measure to take into account the
relative success of each state, based on the logic that states looking to improve their performance
in the area of limiting youth access are more likely to imitate states that have a better track record
than they do. For example, a state with a noncompliance rate of 20% may feel it has much to
learn from a state with a noncompliance rate of only 10%, but little to learn from a state with a
higher noncompliance rate of 40%.
To create Synar Noncompliance Ratio, we take the proportion of minors who were able
to purchase cigarettes during compliance tests in State A and divide this by the proportion of
minors who were able to purchase cigarettes during compliance tests in State B. Thus, this
measure captures the success of State B relative to State A. Values greater than 1 for this
variable indicate that State B has had relatively greater success at keeping tobacco out of the
hands of children. In 2001, for example, minors in Connecticut were able to overcome legal
obstacles and purchase cigarettes 18.1% of the time, while minors in New Jersey obtained
cigarettes 24.6% of the time. The Synar Noncompliance Ratio value for New Jersey in the New
Jersey-Connecticut dyad in 2001 is thus 1.359 (0.246/0.181), indicating that Connecticut was
17
more successful than New Jersey in limiting the ability of minors to purchase cigarettes and
implying that New Jersey could potentially learn from Connecticut’s greater success. The data
for Synar Noncompliance Ratio are available starting in 1997.13
Our second measure of success, Youth Smoking Rate Ratio, is based on the proportion of
high school students who report that they smoked within the past thirty days. Again, this value
compares the relative rate of success in the two states in a dyad, with the rate in State A divided
by the rate in State B. In Michigan in 1999, for example, 34.1% of high school students reported
having smoked within the past thirty days, compared to 40.3% of students in Ohio. Thus, the
value of Youth Smoking Rate Ratio for Michigan in 1999 is 0.846 (.341/.403); since this value is
less than 1, it indicates that Michigan may have less to learn from Ohio in terms of reducing
youth smoking rates than it would from a state with a lower youth smoking rate.14 Because these
surveys are not conducted in each year in each state, we interpolate the missing values. We do
so by inserting the most recent survey results for each year and for each state, as these would be
the best data available to policymakers in other states at the times of their decisions.15
Positive coefficients on these success measures would support the Learning Hypothesis.
Beyond this initial hypothesis, our first set of conditional hypotheses holds that learning from
success may depend on the political expertise within each state’s legislature. To measure
political expertise, we rely on a standard measure, Squire’s (2007) Professionalism Index. This
13 An alternative form of this success variable, which looks at simply the overall levels in State B, rather than the ratios between States A and B, shows additional evidence of learning from more successful states, consistent with the results reported below. 14 An alternative independent variable, using just the level of smoking in State B (i.e., not relative to the level in State A) produces substantively similar results. In addition, capturing the relative smoking rates of high school students in State B to adults in State B, thus potentially accounting for youth access policy success while controlling for the overall rate of smoking in the state, led to substantively similar results as those reported below. 15 Because policymakers cannot predict future smoking rates, it is unsurprising that a linear interpolation between over-time observations in each state performs worse than our measure.
18
measure compares each state legislature, at various points in time, to the U.S. Congress, along
each of three dimensions – legislators’ salaries, the number of days in session per year, and the
number of staffers – and then creates a single index to capture this comparison. States with
higher scores, such as Wisconsin, New York, and California, are considered to be more like
Congress and thus more professionalized, whereas states with lower scores, such as Wyoming,
North Dakota, and New Hampshire, are less professionalized. To test the Political Expertise
Hypotheses, we interact our success measures with Legislative Professionalism in State A.
Our next hypotheses, the Policy Expertise Hypotheses, hold that the learning effect is
conditional on the degree of experience that states have had with the policy area. To capture
such policy expertise, we generate Local Adoptions, measured as the population in cities within
the state that are covered by local youth access laws, divided by the total population of the
state.16 Again, we interact our success measures with this State A variable to test whether the
effect of policy success elsewhere is contingent on expertise.
We also include two other control measures for State B. Because larger and wealthier
states are more likely to be seen as leaders (Walker 1969; Grupp and Richards 1975), these
governments are more likely to be imitated than are smaller and poorer states. Hence, we
include Population in State B and Real Per Capita Income in State B (in thousands of inflation-
adjusted dollars), with the expectation that since the policies in these states are more likely to be
imitated, these variables will yield positive coefficients. We also include our modifying
variables – Legislative Professionalism and Local Adoptions – as controls.17
16 We obtained data on local antismoking laws for all cities with populations above 50,000 from the American Nonsmokers’ Rights Foundation’s Local Tobacco Control Ordinance Database. 17 We include these variables both because they will be necessary for our interactions and to ensure consistency across the results that we report. Our results and conclusions remain
19
Independent Variables: Interstate Relational Characteristics
Beyond the conditions in State B that might entice State A to adopt its policies, we also
control for the relationship between states.18 The classic story in the policy diffusion literature is
about geography, with policies spreading from one neighbor to the next (Walker 1969). We
capture this relationship through the variable Neighbors, which takes a value of 1 if the two
states in the dyad share a common border, and 0 otherwise.
Ready access to information about the policies (and their implications) in neighboring
states makes geographic learning an easy point of external policy focus. Indeed, following the
seminal work of Berry and Berry (1990), geographic neighbor adoption became practically
synonymous with policy diffusion. More recently, however, scholars have moved beyond
geographic diffusion to explore whether ideological, demographic, or other similarities have
contributed to the spread of policies (e.g., Grossback, Nicholson-Crotty, and Peterson 2004;
Volden 2006). While some of the adoption of similar policies by similar states may be
independent of policy diffusion (Volden, Ting, and Carpenter 2008), or may be due to other
mechanisms of diffusion such as economic competition (Berry and Baybeck 2005), similar state
adoption is at least consistent with learning-based diffusion.
To capture political similarity, we create Same Unified Government, which takes on a
value of 1 if both states have unified Republican government or both states have unified
Democratic government, and Ideological Difference, which is the absolute value of the
substantively the same if we exclude these variables (from the non-interactive regressions) or include only the variable that is a constituent term in the interaction. 18 Because our analysis is conditioned on State A already adopting some policy change, the inclusion of variables that influence whether State A adopts any policy is unnecessary. It is thus unsurprising that including State A variables for such considerations as population, health lobbyists, tobacco lobbyists, tobacco production, percent smokers, real per capita income, and state political circumstances does not have a significant effect on the findings reported below.
20
difference in each state’s government ideology as measured by Berry, Ringquist, Fording, and
Hanson (1998). In addition, we create Difference in Percent Smokers and Similar Production,
where the former is the absolute value of the difference in smoking rates across the two states
and the latter is a dichotomous variable that takes a value of 1 when both states are tobacco
producers or neither are tobacco producers. To the extent that similarity should increase the
likelihood that State A will follow State B and adopt its policies, we expect Neighbors, Same
Unified Government, and Similar Production to have positive coefficients, while Ideological
Difference and Difference in Percent Smokers should have negative coefficients.19 All variables,
along with their sources and summary statistics, are detailed in the Appendix.
Results
We begin our dyad-year event history analysis by testing the Learning Hypothesis. Since
the dependent variable in our first test, Movement Toward State B, is dichotomous, we rely on
logit estimation. In addition, to control for possible lack of independence across observations –
particularly because there may be features of State B that we are not capturing in our models and
that will influence whether other states will imitate it – we cluster by State B, an approach that
adjusts the standard errors to account for any discerned non-independence and for possible
heteroskedasticity concerns.20 To control for further possible temporal dependence in the hazard
rates of adoption of State B’s policies, we include year dummies (Beck, Katz, and Tucker 1998).
19 We also investigated whether there might be an interactive effect between the success variables and the similarity variables. Although there is some evidence of such an interactive effect (e.g., between Synar Noncompliance Ratio and Difference in Percent Smokers, and between Youth Smoking Rates and Ideological Difference), such evidence is limited to only certain ranges of some similarity variables. Somewhat to our surprise, overall, there is no systematic pattern of learning being conditional upon similarity. 20 Other functional forms, such as probit or complementary log-log, yield similar results, as do other ways of clustering (e.g., State B-year, State A-State B dyad, State A, or State A-year.)
21
Also, as mentioned above, we limit the dataset to observations in which State A makes some
policy change, and in which State B has already adopted youth access policies not yet found in
State A, thus presenting an opportunity for emulation (Boehmke 2009). Finally, to account for
further limits on the ability to emulate others once State A has already adopted many youth
access restrictions, we include a Previous Laws Adopted control variable that accounts for the
number of previous youth access policy components already in place in State A.21
Model 1 of Table 1 presents the logit results.22 The key variables of interest in testing the
Learning Hypothesis are the measures of relative success in State B, Synar Noncompliance Ratio
and Youth Smoking Rate Ratio. These baseline results provide strong evidence that states are
more likely to shift their policy profiles toward others that have been successful. Both Synar
Noncompliance Ratio and Youth Smoking Rate Ratio have the predicted positive sign, and their
coefficients are highly significant. In addition, both coefficients are substantively significant.
Holding all other variables at their mean, the predicted probability that Movement Toward State
B will take on a value of 1 increases from 0.38 when Synar Noncompliance Ratio is one standard
21 Including Total Previous Laws is especially important as a control, because having already adopted several youth access laws may have a dampening effect on the opportunity to emulate others. This variable could also, in some ways, be thought of as an additional form of policy expertise, beyond local youth access adoptions. Unfortunately, exploring whether the adoption of previous laws leads to greater learning from success elsewhere also raises some serious methodological issues. For example, while more policy experience at the state level may enhance the ability to learn, it also limits the number of new policies about which to learn, as many of those policies most expected to be successful have already been tried. Additionally, the amount of pressure for major policy change may be greater with no youth access laws on the books than with many such laws, leading to a greater likelihood of emulation of successes when there are fewer previous laws adopted. Preliminary analyses show significant nonlinearities in the effect of our success variables on emulation, depending on how many previous laws had been adopted in State A. Further disentangling the interrelated nature of learning from success and internal policy change over time is beyond the scope of our current paper, but is likely to be a fruitful line of future research. 22 For this initial test of the Learning Hypothesis, we temporarily put to the side the notion that variation in political and policy expertise will influence the effects of success.
22
deviation below its mean to 0.51 when it the success measure is one standard deviation above its
mean. Similarly, the predicted probability changes from 0.42 to 0.47 when Youth Smoking Rate
Ratio increases from one standard deviation below its mean to one standard deviation above its
mean; and the probability moves from 0.36 to 0.54 when both success variables move from one
standard deviation below to one standard deviation above their means.23 Thus, comparing the
most successful states (a standard deviation or more above the mean) to the least successful
states (a standard deviation or more below the mean), the policies of the most successful states
are emulated in a significant majority of policy changes whereas the policies of the least
successful states are emulated in less than one third of all policy changes.
[Insert Table 1 about here]
Turning to our control variables, we find no evidence that either Population in State B or
Real Per Capita Income in State B increase the likelihood that State A will imitate State B;
indeed, the latter variable has the opposite sign from what we expected. Further, in looking at
the results for interstate relational considerations, we find little evidence that states are more
likely to follow the lead of similar states, with only Similar Production having the correct sign
and approaching statistical significance. While it was important to control for State B leadership
characteristics and for similarities across states, these sparse statistical patterns indicate that most
of the story about which states are emulated is based on learning from successes, rather than
mere imitation or fads across similar states. The control for previous laws adopted is negative
and significant, as expected. Finally, although the coefficient for Local Laws is insignificant, the
23 We used prvalue in Stata 11 to calculate the predicted probabilities. The size of the coefficients can also be interpreted by their effects on the odds of State A adopting policies that move its profile toward State B. The odds increase by approximately 30% for each additional unit of increase in either of our success measures.
23
coefficient on Legislative Professionalism is negative and strongly significant, suggesting that
more professionalized legislatures are, all else equal, less likely in general to copy other states.
Although Model 1 supports the Learning Hypothesis, there is reason to question whether
these results accurately and completely capture the degree of learning from policy success.
Recall that the dependent variable in this first model is a dichotomous variable, taking on a value
of 1 if a state moves toward another state’s policy profile more than it moves away. Although
this captures the overall relationship between two states’ policy choices, and although it follows
the approach used in other dyadic studies (e.g., Volden 2006), it ignores relevant information.
Suppose, for example, that Indiana and Iowa had no youth access laws at all in a given year, and
Illinois had eight different laws at that time. Now suppose that in the next year, Indiana adopts
one law that is found in Illinois, while Iowa adopts six of the eight laws that are already in effect
in Illinois. Clearly, Iowa has moved more toward Illinois than has Indiana; yet the dependent
variable in Model 1 treats Iowa and Indiana equivalently. More generally, with more than a
dozen different youth access policies that states can adopt, one state could move only slightly
toward another, or could take many more steps toward that other state.
Model 2 of Table 1 uses this additional information with Amount of Movement Toward
State B as the dependent variable. Because our dependent variable is no longer dichotomous, but
rather can take a range of values, we use ordinary least squares for our analysis.24 The results
again show strong support for the Learning Hypothesis. Both Synar Noncompliance Ratio and
Youth Smoking Rate Ratio are positive and strongly significant (p < 0.01, two-tailed tests). A
one-standard-deviation increase in Synar Noncompliance Ratio is associated with an additional
24 In principle, this dependent variable can take values ranging from -16 to 16. In reality, it actually takes values from -5 to 9 during our time period. Ordered yields substantively identical conclusions about our hypotheses and variables, but at the cost of producing results that are much more complex to interpret. Hence, we report OLS estimates in our tables.
24
0.19 components of State B’s laws adopted by each State A. A similar calculation for Youth
Smoking Rate Ratio shows a movement of 0.08 components toward a more successful State B.
In addition, such effects accumulate over time and through compounded learning as other states
experience success with these youth access policies.
Political Expertise and Learning from Success
The results in the previous section demonstrate that successful states are more likely to be
emulated than states that have demonstrated little policy success. A state is much more likely to
adopt aspects of another state’s policy profile if that second state has relatively low rates of youth
smoking or high rates of compliance with the goals of the Synar Amendment. These findings
contribute to the small but growing set of studies that have investigated whether policies in
successful states are more likely to diffuse. We turn now to our other, more novel, hypotheses.
As discussed earlier, political expertise could have two potential effects on the influence
of success: an increasing effect, in which more professional legislatures are better able to learn
about and adopt policies found in successful states; and a decreasing effect, in which more
professional legislatures have less need to draw upon others’ successes. To test these
hypotheses, we need to determine whether the effects of success vary with the level of
professionalism; and to do that, we interact our success and our political expertise measures.
Thus, we interact both Synar Noncompliance Ratio and Youth Smoking Rate Ratio with
Legislative Professionalism, and add these interaction terms (in separate equations) to Model 2.25
25 We present the results for Synar Noncompliance Ratio and Youth Smoking Rate Ratio separately, both because of high collinearity – the correlation between the two interactive terms is r = 0.53 – and because separate reporting makes it easier to disentangle the independent effects of each success measure when we plot the effects of our regressions. The results remain basically unchanged, however, if we include both success measures along with their interaction terms in a single equation. This is also true for the tests of our Policy Expertise Hypotheses.
25
Table 2 displays the results of our tests of the Political Expertise Hypotheses. The results
using the Synar measure, which are presented in Model 3, show that the interaction term is
positive and highly significant (p = 0.003). Similarly, the results based on smoking rates, which
are presented in Model 4, show that the interaction term is positive and significant, albeit at a
more lenient standard (p = 0.066).26
[Insert Table 2 and Figures 1a and 1b about here]
Figures 1a and 1b, which show the marginal effects of each success measure for different
values of Legislative Professionalism, allow us to better interpret the interactive effects.27 The
two figures tell a very similar story. First, for Synar Noncompliance Ratio we see that the effect
of success in other states is positive for all values of Legislative Professionalism, indicating that
regardless of State A’s level of professionalism, an increase in our success measure causes State
A to move toward State B’s policy profile. Thus, both low- and high-professional legislatures
appear to learn from the policy successes of other states, and are more likely to adopt policies
found in other states that are relatively more successful at making it more difficult for teens to
buy cigarettes. Second, the marginal effect of Youth Smoking Rate Ratio (seen in Figure 1b) is
positive for all values of Legislative Professionalism greater than 0.17. Since the mean value of
this measure is 0.20, this finding indicates that legislatures that are average or above average in
terms of capacity are more likely to adopt policies found in another state if that state has had
26 As mentioned earlier, the data on youth smoking rates are available back to 1991, whereas the Synar Noncompliance Ratio data begin in 1997. Because Models 1, 2, and 3 include the Synar Noncompliance Ratio variable, our analysis for those models begins with data from 1997. On the other hand, Model 4, which does not include the Synar Noncompliance Ratio measure, begins with data from 1991. If we run Model 4 utilizing the same years as Models 1, 2, and 3 (i.e., 1997-2002), our results are essentially the same as, or even stronger than, those presented here. This pattern holds true for all tests presented later in this paper. 27 Plotting the marginal effects of interaction terms shows whether the variables of interest – here, our success variables – are significant across the entire range of the modifying variables (Kam and Franzese 2007). We use Fred Boehmke’s grinter program to plot these effects.
26
success at reducing youth smoking. Third, the positive slope of the marginal effects line
indicates that highly professionalized legislatures are more likely to follow the lead of other
states that are more successful than they are.
More specifically, as Kam and Franzese (2007) demonstrate, a significant coefficient on
the interaction term indicates that the marginal effect of the variable of interest – here, our
success measures – differs significantly at high and low values of the modifying variable – here,
Legislative Professionalism. The results, then, provide strong support for the Increasing Effect
hypothesis, showing that although all legislatures are likely to adopt policy profiles that are more
like those found in successful states, legislatures with higher political expertise, as measured by
their professionalism scores, are significantly more likely to do so. The size of these conditional
effects is quite large, as well. For instance the impact of relative Synar compliance success in
State B is three times as large in affecting the odds of a highly professional State A adopting
State B’s policies as it is impacting a State A with a low degree of professionalism. And for the
measure of success based on smoking rates, the least professional states do not even seem to
have the level of political expertise needed to learn from successes found elsewhere.
Policy Expertise and Learning from Success
The results in Table 2 and Figure 1 clearly indicate that the effect of success increases for
policymakers with greater political expertise. Does the same pattern hold true for policy
expertise? That is, do we find a complementary effect, where states with greater internal policy
experience are more likely to adopt policies that are found in relatively more successful states?
Or do we find a substitution effect, where states can substitute internal policy expertise for
external learning, and where the success of other states therefore has less of an effect?
27
To test the Policy Expertise Hypotheses, we follow the same approach that we used for
the Political Expertise Hypotheses. Our measure of policy expertise makes use of information
gleaned from policy experiments at the local level – namely, we rely upon Local Laws, which is
the percentage of the population that is covered by local-level youth access laws. We interact
Local Laws with both Synar Noncompliance Ratio and Youth Smoking Rate Ratio and again add
these interaction terms to Model 2.28
[Insert Table 3 and Figures 2a and 2b about here]
The results, reported in Table 3 and illustrated in Figures 2a and 2b, provide strong
support for the Complementary Effect hypothesis. In Figure 2a we see that that marginal effect
of Synar Noncompliance Ratio is positive across all values of Local Adoptions, indicating that
the more successful is State B, the more likely it is that State A will adopt the types of policies
found in State B. Figure 2b shows a comparable pattern, although the effect of Youth Smoking
Rate Ratio becomes significant at approximately the mean value of Local Laws for our sample
(0.082). In addition, both figures show that the marginal effect of success is increasing in Local
Laws, although this trend is significant only for Youth Smoking Rate Ratio (p = 0.046). Thus, the
data indicate support for the Complementary Effect hypothesis, showing that states with a greater
degree of policy expertise are more likely to draw upon external learning, and show no evidence
that states substitute internal learning for external learning. The results in Tables 2 and 3 thus
show that as state policymakers increase either their political or their policy expertise, they are
more likely to learn from and emulate successful states.
28 Because of high correlations between Local Laws and our success measures, we were concerned that our results might be affected by multicollinearity problems. To address this, we explored numerous ways to reduce the level of multicollinearity (e.g., calculating Local Laws differently, creating dichotomous or trichotomous versions of the variable, and so on). The results were very robust to all of these sorts of changes.
28
Conclusion
State policymakers face a multitude of decisions every year. When deciding whether to
adopt certain policies, they can rely on their own acquired expertise or try to learn from the
experiences of others. Our analysis shows that states do indeed learn from other states. Upon
observing that another state has achieved success in dealing with a policy area, policymakers in
other states are more likely to move toward that successful state, by adopting similar policy
profiles. More specifically, we found that states that are more successful in reducing youth
access to cigarettes and youth smoking rates are more likely to be emulated than are other states.
The finding that successful states are more likely to be emulated is interesting in its own
right, but it also provides a baseline for a more intriguing and novel set of results. Specifically,
we find that, although states frequently learn from other successful states, this effect is not equal
across all states. First, our evidence strongly suggests that states featuring greater political
expertise – namely, those that have more professional legislatures – are more likely to learn from
successful states and to adopt the policies found in those other successful states. This result has
both positive and normative aspects. Positively, it indicates a very specific, institutional
consideration that influences the way in which some policies – in particular, policies of
successful states – diffuse. Normatively, it suggests that, whatever other virtues they may have,
less professional legislatures may be less likely, or perhaps even unable, to emulate the policy
successes of other states. The notion of state-level experimentation is a fundamental aspect of
our federal system; but, to the extent that some states have institutions limiting the expertise of
lawmakers while others have highly professional legislative institutions, the former may be
hindered in their attempts to benefit fully from experimentation.
29
Second, we also find that policy expertise complements learning. In effect, states with
expertise brought about through internal experimentation are better able to emulate successful
states than are states with little experience. More generally, this question of external learning
versus internal policy expertise is not just a point of academic interest, but rather is one with
important substantive implications. Governments in federal systems are constantly assessing
whether, and to what extent, to decentralize policymaking authority; some degree of
decentralization might unleash learning both within and across states. As we uncover here, states
that have robust internal experiments at the local level are able to acquire the needed expertise to
learn from experiments elsewhere – indeed, it may even be necessary for states to allow local
experimentation of their own in order to draw upon successful experimentation elsewhere.
Uncovering these sorts of effects therefore is important in determining the costs and benefits of
devolution decisions.
Taken together, these findings portray a complex relationship among states and localities,
with states learning both horizontally (from other states) and vertically (from policy adoptions
within their state), with the exact nature of learning dependent on both political and policy
expertise. Moreover, these results need not only apply within American federalism. Within the
comparative politics literature, for example, there is significant evidence emerging of countries
learning from one another’s experiences (Graham, Shipan, and Volden 2008). The research here
indicates that learning in comparative contexts might likewise differ depending on the political
institutions that cultivate expertise among policymakers. For example, politicians in countries
with unitary systems may not gain the policy expertise arising from the subnational experiments
of federalist countries, which serve as helpful precursors to learning from outside the country.
30
References
Beck, Nathaniel, Jonathan N. Katz, and Richard Tucker. 1998. Taking Time Seriously: Time-Series-Cross-Section Analysis with a Binary Dependent Variable. American Journal of Political Science 42(4): 1260-1288.
Berry, Frances Stokes, and William D. Berry. 1990. State Lottery Adoptions as Policy
Innovations: An Event History Analysis. American Political Science Review 84: 395-415. Berry, Frances Stokes, and William D. Berry. 1999. Innovation and Diffusion Models in Policy
Research. In Theories of the Policy Process, Paul Sabatier, ed., Boulder, CO: Westview Press, 169-200.
Berry, William D., Evan J. Ringquist, Richard C. Fording, and Russell L. Hanson. 1998.
Measuring Citizen and Government Ideology in the American States, 1960-93. American Journal of Political Science 42(1): 327-348.
Berry, William D., and Brady Baybeck. 2005. Using Geographic Information Systems to Study
Interstate Competition. American Political Science Review 99 (November): 505-519. Berry, William D., Michael B. Berkman, and Stuart Schneiderman. 2000. Legislative
Professionalism and Incumbent Reelection: The Development of Institutional Boundaries. American Political Science Review 94 (4): 859-874.
Boehmke, Frederick J. 2009. Policy Emulation or Policy Convergence? Potential Ambiguities
in the Dyadic Event History Approach to State Policy Emulation. Journal of Politics 71(3): 1125-1140.
Boehmke, Frederick J., and Richard Witmer. 2004. Disentangling Diffusion: The Effects of
Social Learning and Economic Competition on State Policy Innovation and Expansion. Political Research Quarterly 57(1): 39-51.
Braun, Dietmar, and Fabrizio Gilardi. 2006. Taking “Galton’s Problem” Seriously: Towards a
Theory of Policy Diffusion. Journal of Theoretical Politics 18(3): 298-322. Crain, Robert L. 1966. Fluoridation - Diffusion of an Innovation among Cities. Social Forces
44(4): 467-476. Dent, C. and Biglan, A. 2004. Relation Between Access to Tobacco and Adolescent Smoking.
Tobacco Control 13 (4) 334-338. Fiorina, Morris P. 1994. Divided Government in the American States: A Byproduct of
Legislative Professionalism?” American Political Science Review 88: 304�316. Forster, Jean L. and Mark Wolfson 1998. Youth Access to Tobacco: Policies and Politics.
Annual Review of Public Health 18: 203-235.
31
Gilardi, Fabrizio. 2010. Who Learns from What in Policy Diffusion Processes? American
Journal of Political Science 54 (3): 650-666. Gilardi, Fabrizio, and Katharina Füglister. 2008. Empirical Modeling of Policy Diffusion in
Federal States: The Dyadic Approach. Swiss Political Science Review 14(3): 413-450. Gilardi, Fabrizio, Katharina Füglister, and Stéphane Luyet. 2009. Learning from Others: The
Diffusion of Hospital Financing Reforms in OECD Countries. Comparative Political Studies 42(4): 549-573.
Graham, Erin, Charles R. Shipan, and Craig Volden. 2008. The Diffusion of Policy Diffusion.
Presented at the Annual Meeting of the American Political Science Association, Boston. Gray, Virginia. 1973. Innovation in the States: A Diffusion Study. American Political Science
Review 67: 1174-1185. Grossback, Lawrence J., Sean Nicholson-Crotty, and David A.M. Peterson 2004. Ideology and
Learning in Policy Diffusion. American Politics Research 32 (5): 521-545. Grupp, Fred W., Jr., and Alan R. Richards. 1975. Variations in Elite Perceptions of American
States as Referents for Public Policy Making. American Political Science Review 69(3): 850-858.
Huber, John D., Madelaine Pfahler, and Charles R. Shipan. 2001. Legislatures and Statutory
Control of Bureaucracy. 2001. American Journal of Political Science 45 (2): 330-345. Kam, Cindy D., and Robert J. Franzese, Jr. 2007. Modeling and Interpreting Interactive
Hypotheses in Regression Analysis. Ann Arbor: University of Michigan Press. Karch, Andrew. 2007a. Democratic Laboratories. Ann Arbor: University of Michigan Press. Karch, Andrew. 2007b. Emerging Issues and Future Directions in State Policy Diffusion
Research. State Politics and Policy Quarterly 7(1): 54-80. Kousser, Thad. 2005. Term Limits and the Dismantling of State Legislative Professionalism.
New York: Cambridge University Press. Lantz, Paula M., Peter D. Jacobson, Kenneth E. Warner, Jeffrey Wasserman, Harold Pollack,
Julie Berson, and Alexis Ahlstrom. 2000. Investing in Youth Tobacco Control: A Review of Smoking Prevention and Control Strategies. Tobacco Control 9: 47-63.
Luke, D.A., K.A. Stamatakis, and R.C. Brownson. 2000. State Youth-Access Tobacco Control
Policies and Youth Smoking Behavior in the United States. American Journal of Preventative Medicine 19(3): 180–7.
32
Meseguer, Covadonga. 2006. Learning and Economic Policy Choices. European Journal of Political Economy 22(1): 156-178.
Meseguer, Covadonga, and Fabrizio Gilardi. 2009. What Is New in the Study of Policy
Diffusion? Review of International Political Economy 16(3): 527-543. Mossberger, Karen. 1999. State-federal diffusion and policy learning: from enterprise zones to
empowerment zones. Publius: The Journal of Federalism 29(3):31-50. Ross, Hana, and Frank J. Chaloupka. 2004. The Effect of Public Policies and Prices on Youth
Smoking. Southern Economic Journal 70(4): 796-815. Shipan, Charles R., and Craig Volden. 2006. Bottom-Up Federalism: The Diffusion of
Antismoking Policies from U.S. Cities to States. American Journal of Political Science 50(4): 825-843.
Shipan, Charles R., and Craig Volden. 2008. The Mechanisms of Policy Diffusion. American
Journal of Political Science 52(4): 840-857. Simmons, Beth A., Frank Dobbin, and Geoffrey Garrett. 2006. Introduction: The International
Diffusion of Liberalism. International Organization 60(4): 781-810. Squire, Peverill. 2007. Measuring State Legislative Professionalism: The Squire Index
Revisited. State Politics and Policy Quarterly 7(2): 211-227. Squire, Peverill. 2008. Measuring the Professionalization of U.S. State Courts of Last Resort.
State Politics and Policy Quarterly 8(3): 223-238. Volden, Craig. 2006. States as Policy Laboratories: Emulating Successes in the Children’s
Health Insurance Program. American Journal of Political Science 50(2): 294-312. Volden, Craig, Michael M. Ting, and Daniel P. Carpenter. 2008. A Formal Model of Learning
and Policy Diffusion. American Political Science Review 102(3): 319-332. Walker, Jack L. 1969. The Diffusion of Innovations among the American States. American
Political Science Review 63(3): 880-899. Weyland, Kurt. 2005. Theories of Policy Diffusion: Lessons from Latin American Pension
Reform. World Politics 57: 262-295. Weyland, Kurt. 2006. Bounded Rationality and Policy Diffusion: Social Sector Reform in Latin
America. Princeton: Princeton University Press.
33
Appendix: Variable Descriptions, Summary Statistics, Sources
Variable Description Mean St. Dev. Movement Toward State Ba Dependent variable = 1 if State A in dyad moves
toward State B in more categories than it moves away.
0.438 0.496
Amount of Movement Toward State Ba Dependent variable = Number of categories in which State A moves toward State B minus number of categories in which State A moves away from State B.
0.187 1.92
Synar Noncompliance Ratiob Ratio of the proportion of time minors were successful in attempts to purchase cigarettes in State A to the proportion of time minors were successful in attempts to purchase cigarettes in State B. Higher numbers thus indicate that B is having more success than A.
1.343 0.989
Youth Smoking Rate Ratioc Ratio of the proportion of high school students in State A who have smoked within the past thirty days to the proportion of high school students in State B who have smoked within the past thirty days. Higher numbers thus indicate that B is having more success than A.
1.041 0.400
Population in State Bd State B population (millions) 5.68 6.24 Real Per Capita Income in State Be Average income per resident ($000s) in State B
adjusted for inflation 27.9 4.56
Neighbors Dummy = 1 if two states in dyad share border 0.086 0.280 Same Unified Governmentf Dummy = 1 if both states in dyad have unified
Democratic or unified Republican government 0.096 0.295
Ideological Differenceg Absolute value of difference between State Government Ideology for two states in dyad
28.2 20.6
Difference in Percent Smokersh Absolute value of difference between Percent Smokers for two states in dyad
3.34 2.70
Similar Productioni Dummy = 1 if both states or neither states in dyad produce tobacco
0.554 0.497
Legislative Professionalismj Squire composite index of legislative professionalism
0.204 0.132
Local Lawsk Percentage of population covered by local laws (in cities with population>50,000)
0.082 0.126
Data sources: aConstructed based on National Cancer Institute, State Cancer Legislative Database Program, Bethesda, MD: SCLD. bConstructed by authors based on SAMHSA data. cConstructed by authors based on data from the CDC. dConstructed by authors based on data from U.S. Census Bureau. eConstructed by authors based on data from U.S. Bureau of Economic Analysis. f Constructed by authors based on The Book of the States, various years. gConstructed by authors based on Berry, Ringquist, Fording, and Hanson (1998) data on ICPSR website. hConstructed by authors based on Centers for Disease Control and Prevention data. iConstructed by authors based on U.S. Department of Agriculture data. jSquire (2007). kConstructed by authors based on data from the American Nonsmokers’ Rights Foundation.
34
Table 1: Learning for State Youth Access Adoptions Model 1 Model 2 Learning: Relative Success in State B Synar Noncompliance Ratio
0.258*** (0.090)
0.194*** (0.055)
Youth Smoking Rate Ratio
0.246*** (0.092)
0.194*** (0.068)
State B Leadership
Population in State B
0.011 (0.025)
0.013 (0.019)
Real Per Capita Income in State B ($1000s)
-0.022 (0.021)
-0.015 (0.019)
Interstate Relational Considerations Neighbors
-0.056 (0.212)
0.011
(0.190) Same Unified Government
-0.402* (0.222)
-0.285* (0.151)
Ideological Difference
0.002 (0.003)
0.00002 (0.002)
Difference in Percent Smokers
0.0001 (0.015)
-0.001 (0.014)
Similar Production
0.212 (0.139)
0.214* (0.113)
Additional Controls Legislative Professionalism
-2.407*** (0.487)
-0.918*** (0.292)
Local Laws 0.088 (0.376)
-0.422 (0.258)
Wald 2 330.6*** Adj. R2 0.16 N 1424 1424 Robust standard errors in parentheses, clustered by State B. Constant, year dummies, and previous laws adopted variables are included in regression but with results omitted from table due to space considerations. Model 1 dependent variable is Movement Toward State B; Model 2 dependent variable is Amount of Movement Toward State B. Model 1 uses logit; Model 2 uses OLS. ***p < 0.01, **p < 0.05, *p < 0.1 (two-tailed tests).
35
Table 2: Political Expertise and State Youth Access Adoptions Model 3 Model 4 Learning: Relative Success in State B Synar Noncompliance Ratio
0.146*** (0.050)
Youth Smoking Rate Ratio
-0.058 (0.182)
Political Expertise and Learning Synar Ratio × Leg. Professionalism 0.508***
(0.165)
Smoking Ratio × Leg. Professionalism 1.490* (0.788)
State B Leadership
Population in State B
0.017* (0.009)
0.024 (0.023)
Real Per Capita Income in State B ($1000s)
-0.015 (0.017)
-0.027 (0.020)
Interstate Relational Considerations Neighbors
0.057
(0.126)
0.003
(0.153) Same Unified Government
-0.380*** (0.134)
-0.173 (0.115)
Ideological Difference
0.001 (0.002)
-0.001 (0.002)
Difference in Percent Smokers
-0.001 (0.013)
-0.017 (0.015)
Similar Production
0.217*** (0.067)
0.305*** (0.109)
Additional Controls Legislative Professionalism
-0.707* (0.367)
-2.34** (0.918)
Local Laws -0.681*** (0.181)
-1.23*** (0.338)
Adj. R2 0.14 0.18 N 2694 2070 Robust standard errors in parentheses, clustered by State B. Constant, year dummies, and previous laws adopted variables are included in regression but with results omitted from table due to space considerations. Models are OLS; dependent variable is Amount of Movement Toward State B. ***p < 0.01, **p < 0.05, *p < 0.1 (two-tailed tests).
36
Table 3: Policy Expertise and State Youth Access Adoptions Model 5 Model 6 Learning: Relative Success in State B Synar Noncompliance Ratio
0.209*** (0.045)
Youth Smoking Rate Ratio
0.076 (0.109)
Policy Expertise and Learning Synar Ratio × Local Laws 0.175
(0.132)
Smoking Ratio × Local Laws 1.062** (0.513)
State B Leadership
Population in State B
0.017* (0.009)
0.024 (0.023)
Real Per Capita Income in State B ($1000s)
-0.015 (0.017)
-0.027 (0.020)
Interstate Relational Considerations Neighbors
0.053
(0.127)
-0.005 (0.153)
Same Unified Government
-0.381*** (0.134)
-0.169 (0.115)
Ideological Difference
0.001 (0.002)
-0.001 (0.002)
Difference in Percent Smokers
-0.001 (0.014)
-0.016 (0.015)
Similar Production
0.213*** (0.067)
0.299*** (0.110)
Additional Controls Legislative Professionalism
-0.071 (0.228)
-0.678** (0.301)
Local Laws -0.936*** (0.234)
-2.41*** (0.623)
Adj. R2 0.14 0.18 N 2694 2070 Robust standard errors in parentheses, clustered by State B. Constant, year dummies, and previous laws adopted variables are included in regression but with results omitted from table due to space considerations. Models are OLS; dependent variable is Amount of Movement Toward State B. ***p < 0.01, **p < 0.05, *p < 0.1 (two-tailed tests).
37
0.2
.4.6
Ma
rgin
al E
ffe
ct o
f S
yna
r C
om
plia
nce
Ra
tio o
n E
mu
latio
n
0 .1 .2 .3 .4 .5 .6 .7State Legislative Professionalism
Dashed lines give 95% confidence interval.
Figure 1a: Learning (Synar Compliance) Conditional on Political Expertise
-.5
0.5
11.
5
Ma
rgin
al E
ffe
ct o
f Y
ou
th S
mo
kin
g R
ate
Ra
tio o
n E
mu
latio
n
0 .1 .2 .3 .4 .5 .6 .7State Legislative Professionalism
Dashed lines give 95% confidence interval.
Figure 1b: Learning (Smoking Rates) Conditional on Political Expertise
38
.1.2
.3.4
Ma
rgin
al E
ffe
ct o
f S
yna
r C
om
plia
nce
Ra
tio o
n E
mu
latio
n
0 .1 .2 .3 .4Proportion of State A Population Covered by Local Laws
Dashed lines give 95% confidence interval.
Figure 2a: Learning (Synar Compliance) Conditional on Policy Expertise
-.3
0.3
.6.9
Ma
rgin
al E
ffe
ct o
f Y
ou
th S
mo
kin
g R
ate
Ra
tio o
n E
mu
latio
n
0 .1 .2 .3 .4Proportion of State A Population Covered by Local Laws
Dashed lines give 95% confidence interval.
Figure 2b: Learning (Smoking Rates) Conditional on Policy Expertise
Related Documents
