How Change Happens

by Cass R Sunstein

  As noted, the relevant experiments involve people with clear ideological (rather than partisan) convictions, and there appears to be no clear evidence on the specific question whether the same effects would be observed for party. But in light of the general evidence of partyism, there is every reason to believe that they would. Indeed, an important and related study shows that people will follow the views of their party even when those views diverge from their independent judgments—and that they are blind to the effects of party influence.41

  In the relevant study, people—both Democrats and Republicans—were asked their views about an assortment of political issues. As a result, it was possible to obtain a sense of how members of both parties thought about those issues. Otherwise identical groups were then asked about the same issues, but with one difference: they were informed of the views of party leadership. The effect of that difference was significant. Armed with that information, people departed from the views that they would have held if they had not been so armed. Stunningly, the effect of the information “overwhelmed the impact of both the policy’s objective impact and participants’ ideological beliefs.”42 At the same time, people were blind to that impact; they actually said that their judgments were based solely on the merits, not on the effects of learning about the beliefs of party leaders. Here, then, is clear evidence of the consequences of partyism for people’s judgments—and of people’s unawareness of that fact.

  Gridlock

  How does partyism affect the likelihood of social change? In terms of enacted law, the conclusion is simple: Under circumstances of partyism, legislation will be difficult to enact. At least this is so when one or another party has the power to block initiatives. If legislators themselves suffer from partyism, this conclusion should seem self-evident. And even if they do not—even if they feel no antagonism to members of the opposing party and are fully willing to work with them—constituent pressures should push in this direction. In the United States, partyism has been contributing to a highly unusual degree of inactivity in Congress.

  There is a legitimate question whether gridlock is good or bad. If an active Congress would reduce social welfare, there would be a good argument for an inactive Congress. Social welfare is the guide, not the volume of activity. In the relevant period, activity might produce welfare-reducing change. A blocked national legislature is something to lament only if the result, all things considered, is to diminish social welfare. One issue is whether and to what extent the legislative status quo is wanting; if it is not, new enactments are not so desirable. Another issue is whether new enactments would be improvements; if they would not be, then gridlock is a blessing, not a curse.

  A full account of any particular state of affairs would require a theory of optimal deadlock. This is not the place for any such theory. But it seems reasonable to think that if a nation faces a range of serious problems, if imaginable initiatives would reduce or solve those problems, and if partyism makes it difficult to undertake those initiatives, then something is badly amiss. Under imaginable conditions, all of those assumptions are eminently reasonable.

  Solutions

  My principal goal here has been descriptive rather than normative. It is possible to believe that partyism is growing and real but that nothing should be done about it. But the increase in partyism has produced serious problems for American government. If some change is desirable, and if pressing challenges remain unaddressed, how might institutions respond?

  It is tempting to urge that we should aim at its causes, to the extent that we are able to identify them. That would certainly be the most direct and ambitious response. But James Madison’s words in Federalist No. 10, applied to the related phenomenon of faction, are highly relevant here: “Liberty is to faction what air is to fire, an aliment without which it instantly expires. But it could not be less folly to abolish liberty, which is essential to political life, because it nourishes faction, than it would be to wish the annihilation of air, which is essential to animal life, because it imparts to fire its destructive agency.”43

  With Madison’s caution in mind, we should acknowledge that it would be folly to attempt to abolish partyism. To be sure, the nature and degree of partyism are not static. As we have seen, partyism has increased significantly in recent decades, and it might turn out to be much lower twenty years hence than it is now. But changes of that kind cannot easily be engineered. They are more likely to be a function of an array of social forces, including emerging technologies, invisible-hand mechanisms, and the decentralized decisions of a wide range of private and public actors.

  The real solutions lie not in aiming at the causes of partyism but in working to counteract its effects. Consider three possibilities.

  Timing Is Everything

  For obvious reasons, partyism is likely to be most intense before a presidential or midterm election. At that point, negative campaigning will be heightened, and politicians might well be at risk if they attempt to make common cause with those from the opposing party. By contrast, partyism is likely to be reduced in the immediate aftermath of a presidential campaign, when the newly elected commander in chief enjoys a “honeymoon period.” The term is a good one, because it captures a central feature of the immediate aftermath of an election, which is that a new relationship is created with a kind of warm glow. In the presence of that glow, partyism is diminished, at least for a time, and it may be possible to accomplish a great deal.

  The point suggests the immense importance of the period of presidential transition and the need for a president-elect to focus carefully on the top priorities of her or his first term. Clear identification of those priorities, alongside a strategy for bringing them to fruition, has long been exceedingly important. But under conditions of partyism, it is essential to any president-elect—and potentially to the nation as a whole.

  Precommitment

  Under creatively designed laws, significant reform can happen as a result of congressional inaction. Consider, for example, the Defense Base Realignment and Closure Act of 1990,44 which enables the president to appoint the nine members of a base-closing commission. The commission produces a list of recommended military base closures, and if the president approves, they happen—unless Congress enacts a resolution of disapproval within forty-five days. If Congress does nothing, the closures go into effect.

  A more controversial example is known as the sequester.45 In 2011, Congress and President Obama completed a difficult negotiation by agreeing that unless Congress enacted new legislation, automatic (and aggressive) spending cuts would go into effect in 2013.46 At the time, few people favored the automatic cuts; they saw them as a mechanism to force Congress to do its job. But the sequester did go into effect, and for better or worse, it has had major effects on federal spending. The power of the 2011 decision was that it established a drastic outcome if Congress failed to act. The noteworthy surprise was that as a result of partyism, the default outcome actually went into effect.

  If the goal is to improve infrastructure, reform Social Security, make significant changes in fiscal policy, or achieve any other large-scale reform, it is possible to imagine a strategy of this kind. With or without the help of a commission, Congress could allow specified changes to occur on a specified date unless a future Congress says otherwise. Of course there is a serious challenge to efforts of this kind: solutions to the problem of partyism might be defeated by partyism. But in some cases, some kind of precommitment strategy or an alteration of the status quo has sufficient appeal to be feasible.

  Delegation, Change, and Technocracy

  In many cases, the best response to partyism lies in delegation, and in particular in strengthening the hand of technocratic forces within government. I am aware that this conclusion will strike many people as jarring. In response, I suggest that the resolution of many political questions should not turn on politics, at least not in any simple or crude sense. Partyism is unhelpful because partisan differences are irrelevant or nearly so. What m
ost matters are facts, and facts do not turn on political affiliation. Consider these problems:

  1. Should EPA reduce the permissible level of ozone in the ambient air from seventy-five parts per billion (ppb) to 70 ppb, 65 ppb, or 60 pbb?

  2. Should OSHA issue a new rule to control exposure to silica in the construction industry?

  3. Should the Department of Transportation require rearview cameras to be installed in new automobiles?

  4. Should the FDA ban asthma inhalers that emit CFCs?

  All of these questions are highly technical. They cannot possibly be answered without careful engagement with empirical issues. Policymakers need to know the benefits of imaginable policies in terms of health and safety. They also need to know the costs, monetary and otherwise. Would a new rule for silica cost $100 million, or $500 million, or $1 billion? What would be the consequences of those costs? Would they result in fewer jobs or in reduced wages? What are the actual harms associated with exposure to silica at various levels? With proposed regulations, how many lives would be saved?

  To be sure, judgments of value may turn out to play a role in controversies of this kind, but with imaginable empirical projections there may be sufficient consensus to ensure agreement on particular outcomes, even amid significant differences in value and across party lines. If, for example, a silica regulation would cost $1 billion and save merely two lives per year, few people would support it, whatever their party affiliation. And if it would cost $100 million and save seven hundred lives per year, few people would reject it. In any event, it is hopeless to try to answer many of the central questions by reference to one’s party identification.

  No one denies that Republicans and Democrats have different attitudes toward the EPA and OSHA, and those different attitudes might well lead to disagreements about particular initiatives. What I am urging here is that many disagreements are not really about values or partisan commitments but about facts, and when facts are sufficiently engaged, disagreements across party lines will often melt away.

  At least this is so when technocrats, not rigidly bound to ideological convictions, are involved. Many social reforms call for close attention to matters of fact, and questions of value turn out to be secondary. What I am suggesting here is that to choose among reforms, it makes sense to give authority to those who know how to resolve disputes of fact.

  The Fragility of Institutional Judgments

  Although the main goal here has been descriptive rather than normative, we have seen that an understanding of the problem of partyism fortifies the case for certain forms of executive action and for receptivity to a degree of discretion on the part of the executive branch (at least if it is in reasonable hands). It is important to acknowledge, however, that, in practice, people’s judgments about the authority of the executive are greatly and even decisively affected by their approval or disapproval of the incumbent president. Under a Republican president, Democrats do not approve of the idea of a discretion-wielding chief executive, enabled by deferential courts. Under a Democratic president, Republicans tend to have, and even to voice, the same cautions and concerns.

  In this respect, some of the most important institutional judgments are fragile and even unstable. They are weakly held in the sense that they predictably “flip” with changes in the allocation of political power. We could even see institutional judgments as victims of partyism itself. Questions of institutional authority are, in a sense, overwhelmed by short-term assessments of the particular people who are currently occupying relevant offices. For this reason, it is possible that evaluations of arguments in favor a receptive approach to presidential power in light of partyism will be dominated by one factor: evaluation of the current occupant of the Oval Office.

  The aspiration, of course, is that institutional claims can be evaluated behind a kind of veil of ignorance and that short-term considerations about the immediate winners and losers might be put to one side. For political actors, adoption of a veil of ignorance is extremely challenging because short-term electoral considerations often argue against any such veil. If, for example, a Republican politician argues for acceptance of presidential discretion when the president is a Democrat, she might seriously endanger her political prospects. Even for observers, the challenge is real because short-term political considerations have such salience.

  Partyism is real, and it is increasing, and it has serious adverse effects both in daily life and in the political domain. It makes governance more difficult and in some cases even impossible. Even when legislators are aware that a bipartisan agreement would be sensible, they might well be under severe electoral pressure not to enter into it because they might face some kind of reprisal from constituents or colleagues.

  Even under current conditions, the effects of partyism have been far more serious in some periods than in others. On the eve of a midterm election, for example, those effects are likely to be heightened. In the six months after a presidential election, they are likely to be reduced. But for structural reasons, large-scale reductions in partyism are unlikely, certainly in the short term.

  Is this a problem? If the status quo is pretty good and if further action from the national government would likely make things worse, then there would little reason to lament the existence of partyism. In such circumstances, partyism might turn out to be a valuable safeguard. But if a nation faces serious problems and if imaginable initiatives would helpfully address them, then partyism might turn out to create significant dangers for both peace and prosperity.

  At least in the immediate future, it seems unlikely that the United States will be able to make significant progress in reducing the causes of partyism. If such reductions are to occur, they probably will be a product of spontaneous forces rather than of any kind of self-conscious design. The best hope lies in reducing partyism’s effects. I suggest that the most promising approaches include precommitment strategies and a renewed emphasis on the use of technocratic expertise.

  Notes

  1. Shanto Iyengar, Guarav Sood, & Yphtach Lelkes, Affect, Not Ideology: A Social Identity Perspective on Polarization, 76 Pub. Opinion Q. 405 (2012), http://pcl.stanford.edu/research/2012/iyengar-poq-affect-not-ideology.pdf.

  2. Id.

  3. Id.

  4. See, e.g., Anthony Greenwald, Debbie E. McGhee, & Jordan L. K. Schwartz, Measuring Individual Differences in Implicit Cognition: The Implicit Association Test, 74 J. Personality & Soc. Psychol. 1464 (1998); N. Sriram & Anthony G. Greenwald, The Brief Implicit Association Test, 56 Experimental Psychol. 283 (2009) (“In eleven years since its introduction, the Implicit Association Test … has been used in several hundred studies to provide measures of association strengths.”).

  5. E.g., Greenwald, McGhee, & Schwartz, supra note 4, at 1474; Scott A. Ottaway, Davis C. Hayden, & Mark A. Oakes, Implicit Attitudes and Racism: Effects of Word Familiarity and Frequency on the Implicit Association Test, 19 Soc. Cognition 97, 130 (2001); Shanto Iyengar & Sean J. Westwood, Fear and Loathing across Party Lines: New Evidence on Group Polarization, 59 Am. J. Polit. Sci. 690 (2015).

  6. Iyengar & Westwood, supra note 5.

  7. See Iyengar, Sood, & Lelkes, supra note 1, at 416 (showing a steady decrease in racial polarization from 1964 to 2008).

  8. Id. at 415–418.

  9. Id.

  10. Id.

  11. Paul Taylor et al., The Rise of Intermarriage, Pew Social & Demographic Trends 7 (February 16, 2012), http://www.pewsocialtrends.org/files/2012/02/SDT-Intermarriage-II.pdf.

  12. Id. at 36.

  13. Frank Newport, In U.S., 87% Approve of Black-White Marriage, vs. 4% in 1958, Gallup (July 25, 2013), http://www.gallup.com/poll/163697/approve-marriage-blacks-whites.aspx.

  14. Iyengar & Westwood, supra note 5.

  15. Id. at 15–16.

  16. Id. at 16–17.

  17. Id. at 18.

  18. Geoffrey D. Munro, Terell P. Lasane, & Scott P. Leary, Political Partisan Prejudice: Selective Di
stortion and Weighting of Evaluative Categories in College Admissions Applications, 40 J. Applied Soc. Psych. 2434, 2440 (2010).

  19. Id. at 2444–2445.

  20. Iyengar & Westwood, supra note 5.

  21. Yphtach Lelkes & Sean J. Westwood, The Nature and Limits of Partisan Prejudice (Working Paper, 2014).

  22. Id. at 9.

  23. Id. at 10.

  24. Id. at 11.

  25. Id. at 14.

  26. See Lilliana Mason, “I Disrespectfully Agree”: The Differential Effects of Partisan Sorting on Social and Issue Polarization, 59 Am. J. Pol. Sci. 128 (2015); Adrian Furnham, Factors Relating to the Allocation of Medical Resources, 11 J. Soc. Behav. & Personality 615, 620 (1996).

  27. See Iyengar, Sood, & Lelkes, supra note 1, at 422–423; Mason, supra note 26.

  28. See Iyengar, Sood, & Lelkes, supra note 1, at 425–427 (finding that residence in a battleground state during an election year correlates significantly with intensity of partisan affect and that partisan affect increases significantly over the course of a campaign, especially in battleground states); Guarav Sood, Shanto Iyengar, & Kyle Dropp, Coming to Dislike Your Opponents: The Polarizing Impact of Political Campaigns (Working Paper, April 2013) (finding that over the course of a campaign, partisans form more negative views of the opposing party, and the most strongly correlated feature is exposure to televised political advertising, especially negative ads).

  29. See Yphtach Lelkes, Shanto Iyengar, & Gaurav Sood, The Hostile Audience: Selective Exposure to Partisan Sources and Affective Polarization (Working Paper, 2013) (finding that for partisans who pay attention to politics, cable access is correlated with greater partisan affect in years when cable carries partisan content, and further finding that the preference of partisans for choosing, between MSNBC and Fox News, the news sources amenable to their party “in and of itself—is sufficient to predict partisan animus, greater affect for in-party elites vis-à-vis out-party elites, greater social distance between partisans, and a preference for attack-oriented campaign rhetoric”).