With the publication of this book, Pippa Norris further secures her reputation as one of our leading scholars of democratization and democratic consolidation. But far from being cloistered in the academy, Norris has also been at the forefront of international policy debates over democracy promotion, having recently served as director for democratic governance at the United Nations Development Program. In Driving Democracy she has sought to bridge both worlds and has succeeded brilliantly. As a consequence, this is a book that most political scientists, irrespective of subfield, can and should read with profit.
The big question that Norris asks is the one that appears as the book's subtitle: “Do power-sharing institutions work” better in new democracies characterized by deep social divisions than the power-concentrating alternatives? More precisely, she asks whether democracies that have constitutional arrangements and formal institutions that constrain executive power achieve higher democracy scores (e.g., from Freedom House and Polity) than those in which political power is more concentrated in the executive branch. Her answer, based on an analysis of a newly constructed data set of all democracies from 1973 to 2004, is clearly “yes.”
Norris's interest in power-sharing institutions derives from the work of Arend Lijphart and the followers of consociational theory more generally. Briefly, Lijphart's theory was inspired by the case of the Netherlands, which had achieved a stable democracy despite being fragmented by religious and economic differences. He argued that Holland's success was due to its rejection of Westminster democracy in favor of a system that gave many political parties access to power, along with a set of formal institutions that required those parties to cooperate in the formation of a government. The leaders of the country's different social groupings had no choice but to reach out beyond their own power base if they were to build a viable political coalition. Conversely, each small center of power recognized that it played a key role in making the political system work as a whole. In short, just as the Dutch worked together in order to keep out the seawater, so too did they design a political system—a “consociational democracy”—that depended upon intensely cooperative behavior.
In considering whether power-sharing institutions are generally associated with higher levels of democracy in the world's youngest democratic states, Norris adopts a multimethodological approach, combining large-n research with case studies and policy analysis. In particular, she finds that in ethnically divided societies, of the kind that exist throughout much of Africa, for example, power-sharing regimes achieve far higher democracy scores than do their power-concentrating rivals. It is this research that leads her to conclude (p. 223) that “power-sharing constitutional arrangements … contribute toward lasting peace settlements and sustainable democracy.”
As noted, Norris comes to this position on the basis of both quantitative and qualitative research. First, the heart of the book is a series of regression models in which she runs each significant institutional component of a power-sharing regime (e.g., electoral systems, degree of decentralization, extent of a free press, and so forth) against a series of variables in an effort to detect whether democratic performance is higher or lower where these regime attributes are present. She finds that proportional representation, less executive power, decentralization, and a free press are all powerfully associated with higher levels of democracy. Second, she then examines a number of paired country case studies (e.g., Togo vs. Benin; Ukraine and Uzbekistan) where similar findings emerge. Finally, she looks at public policy and particularly the practices of the international community in promoting democracy, again making the case for power-sharing regimes.
The book is both a tour de force and a tour d'horizon. Norris seems to know pretty much everything about the world's youngest (and indeed oldest) democracies and how their formal institutions operate; one could not hope for a better guidebook through this thicket of issues and places. She is also refreshingly honest about the limitations of both her theoretical model (consociationalism) and her empirical work. All this makes for a book that is difficult to criticize, and the following comments represent quibbles more than substantive disagreements.
First, I believe that the book would have been even stronger had it been organized slightly differently. After a useful introduction that provides an overview of her argument, Norris has two methodological and empirical chapters that might have been better placed in the appendix. She then takes a chapter to analyze the relationship between wealth and democracy, nicely summarizing the literature but adding little that is novel. In short, a reader could easily move from the introduction to Chapter 5 without missing much of significance.
Second, there is an underlying tension in this book—which Norris fully recognizes—between its large-n regressions and the fine-grained case studies and public policy analysis that she also provides. The problem facing every academic who wants to speak to contemporary policy issues concerns how to apply broad-brush theoretical and empirical findings to the very particular policy concerns of specific public officials located in space and time. The question facing the policymaker is not whether power-sharing regimes are good in general, but whether they should be applied in country A that is just emerging from a civil war. The answer, of course, will reflect not only our ideas about what works and what does not but also the facts on the ground with respect to those who hold power and what they want in any given situation.
This issue speaks to a third concern. Despite Norris's compelling line of argument, the fact is that at least some of the power-sharing institutions that she recommends—like federalism or decentralization more generally—often have ambiguous political and economic outcomes, both in the developing and industrial worlds. Scholars and international civil servants, for example, have found that the oft-touted benefits of decentralization—which itself was pressed upon many countries by the World Bank, among other organizations, for the very reasons that Norris provides—have proved elusive in any number of settings. Rather than diffusing power, decentralization has instead sometimes strengthened the power of a local chief and his (less frequently her) followers at the expense of other social groups. As a consequence, corruption has increased following several decentralization episodes, as outsiders must now pay insiders for access to public services. In this context it is useful to recall the ways that the consolidation of the U.S. national state in the late nineteenth century challenged the powers of the big-city mayors who similarly once ruled their fiefdoms to the near-exclusive benefit of core supporters. This is not to say that federalism and decentralization are bad, but that the jury is still out regarding their political and economic performance (and, of course, we have some major social science experiments now ongoing in Afghanistan and Iraq along these lines).
Fourth, Norris never fully addresses the (very big) question of the extent to which political outcomes are endogenous to preexisting social relationships, which themselves may reflect even deeper geographic and historic forces. She tries to get at this issue through her paired country cases, but although these are illuminating, they also serve to remind us of how unique each country setting is at the end of the day. And while she highlights the role of ethnic fragmentation and class differences in explaining her preference for power-sharing regimes, she does not address, for example, the way in which a country's geography and early political development might influence its subsequent set of possible political arrangements. (While she does mention the natural resource curse, she says nothing, for example, about agriculture and only a little about urbanization; but the distribution of land and the population shift from land to city are, of course, of major importance to political history.)
Finally, I wish Norris's concluding chapter had provided more in terms of directions for future research and lessons for policymakers. As students are now wont to ask, “what are the take-aways” for academics and public officials? To be sure, the author has made a convincing case on behalf of the democratic virtues of power-sharing regimes. But her final chapter basically recapitulates her argument, and addresses potential criticisms of it, more than it addresses fresh questions that others should address.
Still, these criticisms are minor in comparison with all that Norris has accomplished in this volume. Driving Democracy exemplifies policy-relevant scholarship of the highest order. It takes on a big topic and uses a multimethodological approach in order to reach important findings. The book is also a pleasure to read. For all these reasons, it deserves an audience that goes way beyond the field of comparative politics.
