Working With The Grain: an important new book on rethinking approaches to governance

Even though it’s relatively short (223 pages), Working With the Grain (WWTG) took me two months to finish, but I’m glad I did. It adds to a growing and significant bodywwg-cover of literature on ‘doing development differently’/’thinking and working politically’ – Matt Andrews, Adrian Leftwich, David Booth, Diana Cammack, Sue Unsworth etc. (Like Matt and Adrian, WWTG author Brian Levy is a white South African – what attracts that particular group to rethinking governance would make an interesting study in itself.)

Brian summarizes the common elements of this emerging school of thought as:

‘An insistence that the appropriate point of departure for engagement is with the way things actually are on the ground — not some normative vision of how they should be;

A focus on working to solve very specific development problems – moving away from a pre-occupation with longer-term reforms of broader systems and processes, where results are long in coming and hard to discern;

Recognition that no blueprint can adequately capture the complex reality of a specific setting, and thus that implementation must inevitably involve a process of iterative adaptation.’

What makes this book special is Brian’s CV – two decades at the World Bank, which experience he raids to provide great case studies throughout. It feels like he’s now gone back into academia (he teaches at Johns Hopkins and the University of Cape Town) partly to make sense of what he’s learned from 20 years of success and (more often) failure (he characterizes the orthodox governance approach as ‘a breathtaking combination of naivete and amnesia’). Unlike most such tomes, I found it clearer on the ‘so whats’, than the general diagnostic, which tends to get bogged down in endless 3 point lists and typologies (hence the two months).

Levy sees different countries’ development paths as emerging from the interaction of economic growth and institutions. Growth transforms politics by creating a vocal middle class/ business sector. Occasionally, growth and institutions co-evolve, as in the US in the 19th Century. At other times institutions lead (South Africa) or growth leads (South Korea). That raises interesting questions about NGOs’ focus on the masses (social movements, active citizens etc) – does that mean that we both underestimate the importance of the rising middle classes and underplay the politically transformatory impact of growth?

WWTG comes up with one of those 2×2 typologies so beloved of social scientists to describe different regime types along two dimensions of governance – whether their polities are dominant or competitive; within each of these, whether the rules of the game centre around personalized deal-making or the impersonal application of the rule of law. This generates four broad country types:

Dominant discretionary, where strong political leadership (perhaps military, perhaps civilian; perhaps organized around a political party, perhaps a charismatic individual) has successfully consolidated its grip on power, but formal institutions remain weak, so rule is personalized. [examples: Korea under General Park Chung-hee; Ethiopia under Meles Zenawi]

Rule-by-law dominant, where institutions are more impersonal, but political control remains monopolized. [Korea from late-1970s through to its transition to democracy; contemporary China; perhaps Morocco, Jordan etc]

Personalized-competitive, where politics is competitive, but the rules of the game governing both the polity and the economy remain personalized. [Bangladesh, Zambia and many other low-income countries which democratized in the 1990s]

Rule-of-law competitive, where the political and economic rules have become more impersonal – though some other necessary aspects of democratic sustainability have not yet been achieved. [South Africa]

Over time, the key to development is how institutions emerge and evolve in each quadrant, and change the nature of the system.

The table below shows his suggestions for the kind of reforms that might actually ‘work with the grain’ in different regime types. He has a clear preference for incrementalism, especially when a country is on a positive growth path that is likely to generate positive pressures for reform in due course. He is also clearly thinking from the vantage point of a major donor in dialogue with national governments – some substantial rethinking would be required to make this relevant to smaller players like NGOs.

Levy options tableThere are some other big plusses, including some great history. Your starter for 10 – which dysfunctional state is being described here? [Answer at the bottom of the post]

‘All federal positions were patronage appointments – either made directly by the President, or allocated by him among members of congress to distribute as ‘spoils’. Federal employees had no job tenure; they were removed upon the defeat of their political benefactors. Employees were required to be politically active on behalf of their sponsors, and to contribute 2-10 percent of their salaries to party coffers – else risk losing their job.’

I also liked the honest agonizing and discussion of the dilemmas Levy has faced in his time at the Bank – when is it OK to endorse corruption and/or authoritarianism? How come hopeless political systems like Bangladesh can generate so much social progress? This is all good, grown up, reflexive stuff.

The style is, to be honest, a mixed bag (which itself is a marked improvement on a lot of other governance authors). Sometimes he writes with real verve – ‘‘Advocacy of democracy has been maximalist, shrill, its pendulum swinging repeatedly from euphoria to outrage, with disappointment along the way.’

But then try getting to the end of this paragraph without losing the will to live:

‘In principle, the relevant principals can comprise both governmental and non-governmental actors – although including governmental actors as a co-equal among the principals presumes that, in the context of the specific collaborative endeavor being considered, they have similar standing as co-principals as do non-governmental actors.’

Overall, the book is subtle, complex, often confusing and repays careful study. The complexity is born of deep first hand knowledge, and in the end, suggests that for all Levy’s heroic attempts to distil some general lessons, the roots of success and failure are really only visible with hindsight. Governance advisers and reformers will continue to flounder in the fog, but WWTG, and the other revisionist books and papers, can help a little by discounting some of the bad ideas, and maybe showing people how to look for (and recognize) success a little earlier. That feels like a worthwhile contribution.

More info on the WWTG website. Answer  to the Horrible History question: the United States for much of 19th Century

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


3 Responses to “Working With The Grain: an important new book on rethinking approaches to governance”
  1. Brian Levy

    Thanks, Duncan, for drawing attention to Working with the Grain. The FP2P review points towards the two (perhaps not entirely compatible) goals which I had in writing the book. One of my goals was to write, anchored in my lived experience, an accessible tour d’horizon of the current, cutting edge of development thinking and practice — and how it got that way. I’m pleased that you liked that part of the book – and that you felt it, indeed, offers a thoughtful, honest insider’s account of the challenges of integrating governance and inclusive growth into development practice.

    My second goal was to provide an analytically robust conceptual framework as a foundation for moving that thinking forward. The book brings together a variety of theoretical contributions which rarely are considered in an integrated way. As anyone who has labored through the work of Mushtaq Khan, Douglass North, Elinor Ostrom and Oliver Williamson knows, this involves a lot of new terminology (= jargon?) – but the payoff can be high. I have tried to make this material more accessible than usual – and will leave it to readers to judge whether I’ve succeeded in that goal.

    More fundamentally, I have tried to use the analytical concepts to help push the analytical foundation of development practice beyond the tired polarities of Bill Easterly’s best practice, technocrat ‘tyrant experts’ and his bold ‘searchers’, plunging, gloriously free, into the unknown. As the FP2P review highlights, I use the book’s conceptual platform to identify four distinctive country-types – each characterized by distinctive incentives and constraints to development policymaking and implementation. The book explores in depth how both reform priorities and effective approaches to implementation vary radically and systematically across the country-types. The aim is to give content to the notion of “good fit”— directing attention away from off-the-shelf blueprints, and hopefully laying out a practical, analytically grounded set of options that can help us engage constructively with the governance ambiguities of our early 21st century world.

  2. William Ferguson

    I hope to continue the dialogue initiated by Duncan’s insightful post. First, I concur with his evaluation of the usefulness of WWTF’s discussion of numerous cases and histories. Hence I will restrict my comments to the theoretical side of WWTF’s argument. Here I confer with Brian’s post on the usefulness of WWTFs basic analytical categories. Echoing his statement concerning the ubiquity of collective action, I will focuses my comments on how the framework outlined in WWTF offers a useful foundation for analyzing collective-action problems of development.

    While not primarily a theoretical work, WWTF nonetheless sketches a conceptual framework that is based on a basic typology of development trajectories (with the important caveat that movement is not unidirectional) and a set of corresponding principles related how the strength of institutions and incentives of organizations (and individuals) shape possible dynamic paths. As Brian says, this approach merges ideas developed by various authors, including Elinor Ostrom, Douglass North, and Oliver Williamson. Levy’s approach provides a relatively simple lens through which analysts and policymakers may begin to conceptualize the complex tradeoffs and dynamics of the myriad collective-action problems that accompany development processes (for a comprehensive treatment of the political economy of collective-action, see my recent book, Collective Action and Exchange: A Game-Theoretic Approach to Contemporary Political Economy, Stanford University Press, 2013).

    The WWTF approach, moreover, offers a method for navigating between two contrasting positions: on one hand, a “best practice” or “one-size fits all” approach to development (as in market liberalization combined with democratic reforms); on the other, a fully context-specific, case-by case approach. WWTF’s typology can balance basic principles of application with critical features of pertinent political-economic contexts.

    WWTF’s Figure 2.1 shows the basic typology (I had hoped to copy that figure here, but I seem to be unable to)

    Several features are worth noting. There are six categories in the diagram: Conflict on left, followed by a 2×2 matrix with vertical categories Dominant Trajectory and Competitive Trajectory and Horizontal categories Personalized and Rule-by Law, and a sixth category, Sustainable Democracy on the right. The categories represent idealized types that may also represent end points of horizontal and vertical spectra. Movement in any direction is possible. This important qualification implies the potential fragility of an extant set of institutional arrangements, with less stability, on average, at personalized end of the horizontal spectrum. In a dynamic framework, these six idealized categories constitute categories of social punctuated equilibria—arrangements that achieve stability over relatively long periods that are interrupted by rapid bursts of change. Such punctuation moves a society to a different cell in the diagram.

    The framework implies two general categories of developmental collective-action problems (CAPs). First, in order to initiate developmental dynamics, a society must somehow exit (or avoid) the conflict equilibrium. The CAPs associated with such escape relate to reducing incentives for powerful organizations (or individuals) to use violence. Such incentives emerge via establishing a political settlement that, in turn, typically involves some mechanism for distributing economic rents. This latter point is, in itself, important since (as also emphasized by Acemoglu and Robinson, 2013) well-meaning attempts to reduce or eliminate rents may have the unintended consequence of generating political instability and thus reducing, rather than increasing, a society’s future growth potential.

    WWTF describes to two basic types of political settlement that address this first set of CAPs: settlements that establish dominant political regimes and ones that create competitive political regimes. Each initiates a distinct developmental trajectories, though movement between the two is possible. Dominant settlements, as the name suggests, establish rule by a single dominant coalition (or person)—whether by a dictator, one-party state, or the military. Competitive settlements rely on an informal or formal set of rules that specify procedures for transferring power between two or more coalitions. WWTF indicates a simple criterion for distinguishing the trajectories: how a disparity in violence potential between a ruling coalition and its opposition affects the level of commitment the opposition would need to displace current rulers. In the case of a dominant political settlement, a large difference in violence potential implies that replacing a ruling coalition would require substantial commitment among opponents, and thus resolution of a complex organizational CAP—typically a remote possibility. A low disparity in violence potential, on the other hand, implies that dislodging a ruling coalition requires a relatively small degree of commitment among opponents—posing a relatively easy organizational CAP. Understanding these conditions, the relevant coalitions negotiate a set of informal or formal rules for transferring power: a competitive political settlement.

    This idealized distinction is important because it points to key power dynamics and prospects for continuity that influence developmental processes. WWTF notes that intermediate cases (e.g., Indonesia) are not uncommon. Moreover, movement back and forth between the two regime categories is not uncommon (e.g., Bangladesh).

    Either type of political settlement may signify resolution of a first-order CAP (resolution associated with reaching agreements, see Chapter 2 of Ferguson, 2013), but under either type of political settlement, subsequent development involves subsequent resolution a complex set of ensuing CAPs associated with three basic conditions:

    1) Establishing credible commitments by powerful parties (private or public) to refrain from seizing the gains from others’ investments in knowledge and physical capital. Note that to generate economic growth, such commitments need not be universal; they could be selective (e.g., Mexico under Diaz between 1870 and 1910). Furthermore, to be credible, such commitments need to attain at least a minimal threshold of both legitimacy and enforceability—the latter posing a second-order CAP (see Chapter 3 of Ferguson, 2013). WWTF usefully notes that credible enforcement requires establishing some form of “threat-trumping” dynamic whereby “protagonists” of development may establish “countervailing influence networks” that restrain potential predators.

    2) Provision of basic public services to support whatever commerce may ensue.

    3) Provision of basic infrastructure for the relevant type of commerce.

    China, for example, has achieved all three conditions with a one-party dominant political regime and thereby generated steady economic growth (and substantial economic development by many other criteria) for over three decades.

    Resolution of the myriad associated CAPs, however, depends not only on the nature of a society’s extant political settlement but also, fundamentally, on the underlying strength of its institutions. Here again WWTF posits two idealized categories (poles of a spectrum): personalized rule and the rule-by-law. In the former case, the arrangements and rules for monitoring and enforcement “are built around the specific identities (and threat potential) of the parties involved (WWTF 20).” Such arrangements, therefore, depend on specific personalities and personal networks and may shift or collapse upon the replacement or death of key individuals. In contrast, a rule-of-law or impersonal order signifies codified rules (laws; formal institutions) with monitoring and enforcement activities assigned to third-party (relatively impartial) organizations. Specific procedures, thus do not depend on specific personalities or personal networks. Accordingly, a rule-by-law (punctuated) equilibrium achieves a type of stability and impartiality that personal rule orders cannot generate.

    Again, this idealized typology suggests fundamental distinctions among the myriad dynamics of developmental CAPs. For example, in the dominant political trajectory, the most important CAPs that accompany establishing a rule-by-law (in both political and economic domains) involve instituting credible restraints on the power of key individuals and organizations within the dominant coalition. After 1979, for example, China developed a set of economic institutions that gradually imposed greater restraints on the ability of specific individuals and organizations within the state and the CCP to seize the benefits of others’ economic investments.

    By contrast, societies with a competitive settlement and personalized rule operate on the “edge of chaos” – itself a useful metaphor that conveys the underlying fragility of institutional arrangements. In this case, potentially binding constraints that can short-circuit development (as in Rodrik 1998) include weak infrastructure, corruption via patronage agreements, alienation of non-elite population, weak rule-by-law, and associated commitment problems. Economic actors, therefore, face prospects of seizure from both private and public parties. In such a system, moreover, the distribution of rents serves as the “currency” of politics via discretionary conferral and withdrawal. Movement to a rule-by-law (in the economic sphere) involves the formalization of property rights (this idea mirrors Heilbroner’s (1992) argument that the historical “economic revolution” in Western Europe involved the emergence of the economic sphere as distinct from the political and cultural spheres). In such cases, WWTF recommends that development policy follow problem-driven iterative adaptation (as in Andrews, 2013) that seeks “islands of effectiveness.” The establishment of the garment industry in fragile institutional environment of Bangladesh in the early 1980s offers an example.

    By contrast, dominant regimes that have established a rule-by-law, especially in the economic sphere, can sometimes achieve successful economic growth, as in post-1979 China and South Korea between the late-1950s and 1987. The latter case, moreover, illustrates the potential for economic development via export-oriented industrial policy orchestrated by a developmental state (combined with aggressive education policy and prior, externally imposed, land reform). In this instance, economic growth initiated a virtuous circle feedback mechanism whereby strengthening of the middle class and civil society generated increasing pressure to establish a democratic political regime. This pressure reached a critical mass in 1987, initiating a democratic transition that has, so far, been sustainable.

    Overall, then, for societies that have escaped a conflict order via some form of political settlement—but which are not sustainable democracies—we see a matrix consisting of four idealized development categories—a typology that is far more sophisticated than Rostow’s (1960) stages of growth. Each cell of WWTF’s Figure 2.1 connotes a basic configuration of institutions and associated incentives for key actors, a type of punctuated equilibrium (with greater or lesser degrees of stability), that implies distinct types of developmental collective-action problems and an associated set of potentially binding constraints. Feasible policy strategies, particularly with respect to difficult problems of implementation, need to account for such distinctions, implied incentives and dynamics, and the potential for movement in any direction. In this regard, the theoretical side of WWTF has offered a real contribution.
    William D. Ferguson
    Gertrude B. Austin Professor of Economics
    Secretary-Treasurer, Midwest Economics Association
    Chair, Economics Department and Policy Studies Concentration
    Grinnell College

    • Brian Levy

      Thanks, William, for this extended comment on the theoretical underpinnings of Working with the Grain. Your recent book, Collective Action and Exchange: A Game Theoretic Approach to Contemporary Political Economy (described by Herb Gintis as “a forceful introduction to the analytic techniques involved in this intellectual revolution”…..) is an extended, rigorous exploration of what I lay out in a much more simplified, intuitive way. So I really appreciate your vote of confidence.