Working With the Grain: Integrating Governance and Growth in Development StrategiesOxford University Press, 2014, 288 pp., $31.50
Looking over trends in international development over the past three or four decades, it becomes clear that almost all country gains have come in spite of the profuse academic literature, which has conspicuously lacked practical grounding. With some notable exceptions, the literature has mostly been either simplistic and anecdotal or laden with abstract theory and ideology. When not theoretical, it has veered wildly toward single-factor explanation and advocacy, leading to confusing policy prescriptions. Such advocacy has often ignored local contexts, particularly the willingness and capacity of a country’s professional class to accept proposed reforms, let alone to support implementation.Instead, the development economics and political science literature of the 1960s and 1970s focused on waging debates between the likes of modernization policy advocates and those who thought dependency on foreign expertise and capital undermined national development. This led to significant policy failures in such countries as Chile and Peru, including land reform, nationalizations, and populist versions of statist policies instituted later by countries like Venezuela.In the 1980s the literature shifted focus to righting macro-economic imbalances and liberalizing markets through such policies as structural adjustment, which typically required state retrenchment, fiscal austerity, and deregulation of industry and the financial sector. Opponents of budget discipline and state rollbacks used quaint terms like “fiscal fascism” to give their causes gravamen. After the fall of the Berlin Wall, the apparent triumph of the Western liberal democratic model led to a set of almost predictable, uniform development policies. Development progress was now to be premised on the notion that good policies and economies required good governance institutions—rule of law, low corruption, political accountability, clean elections, and capable bureaucracies. Since these were the institutional embodiment of Western enlightenment values, who could disagree?Currently, the development policy focus is on “best practices” and how to transfer them from one country to another. Most of the development literature that gave rise to this approach started from descriptions of ideal end states, usually based on Western models. It largely ignored the challenges in getting there, and the frustrating practical journey in the field, working with locals, that may take time and usually require “good-fit” rather than “best-fit” policies. The literature likewise ignored how to get the development process going and to sustain it in different country contexts. Describing well-governed states to officials in the largely failed states of Pakistan or Honduras does not, as Brian Levy writes, “conjure them into existence out of thin air.”Levy’s new book, Working With the Grain, is a welcome alternative to the dominant “best practices” approach to governance and development policy reform: a terse, sober, practical, and positive book, filled with personal anecdotes from the field. Levy, for many years a World Bank economist, holds a joint professorship at the University of Cape Town (South Africa) and Johns Hopkins University (SAIS) in Washington. His approach is incremental rather whole-systems reform, tailored to failing states in conflict zones: developing micro-level initiatives based on in-depth, country-specific knowledge, which can serve as “islands of effectiveness” within a “broader sea of dysfunction.” This approach, Levy argues, can create what Albert O. Hirschman called “virtuous circles”, linkages between governance and growth that expand, including more people and local institutions, and move the country toward development.Drawing on his decades of experience in the field, Levy develops a comparative typology that places countries into four categories, distinguishing them along two dimensions: income levels and governance institutions. It assesses whether polities are competitive or dominant, and within them whether rules of the game are characterized by personalized deal-making or impersonal application of rule of law. This tool allows for more precise analysis of country trajectories, which in turn allows for more detailed, well-tailored policy responses. It also allows the user to explore development options from a dynamic perspective, as opposed to a static description of good governance elements. According to Levy,
Each of the four country types comprises a distinctive platform for development, with distinctive incentives for the participants, distinctive constraints and risks, and distinctive frontier challenges. Careful attention to these incentives and constraints provides a way of identifying specific policy actions that are both worthwhile and feasible, given country-specific institutional realities.
The aim is to create forward momentum so that gains in one sector, such as middle-class growth, might strengthen civil society and the private sector. Applying the typology yields useful insights, for instance that the creation of sustainable democracies requires more than competitive elections. In this context, they require complementary, impersonal check-and-balance institutions that constrain personalized and often abusive exercises of power.How did he develop the typology? Working in Dhaka, Bangladesh, in 2005, he was overwhelmed by the traffic in human suffering—the filthy street children, the abandoned elderly. During the wet season 20 percent of the country was underwater, and corrupt property developers would block drainage in order to drive off former owners, then drain the lands to sell as parcels to the rapidly growing middle class. He studied a country with the best and worst of development records. Bangladesh’s gross primary school enrollment was 98 percent in 2001; power alternated via elections in 1996, 2001, and 2008. But it was rated the most corrupt of 150 countries by Transparency International in 2005, and suffered fraudulent elections in 2006. Bangladesh seemed to violate both the good governance thesis and the private sector-led development thesis on how growth occurs. Observing this, Levy developed a competitive trajectory that included conditions of political competition, personalized bargaining among elites, and clientelism. And he hypothesized that the situation in places like Bangladesh are what one should expect in the early stages of this competitive trajectory.How should development policies look in this confusing context? To get the virtuous circles of development going and sustain them, he argues, requires a “just enough” strategy: address specific capacity and institutional constraints in a kind of “high-wire” crisis management act, tentatively moving things ahead with the expectation of course-corrections. He then explores detailed “with the grain” development strategies that have worked in specific country settings and might work under similar conditions elsewhere. South Korea and Ethiopia are used as examples of elite dominance, which under different conditions achieved poverty reduction and growth. The challenge there was to unlock private investment without inviting excessive elite predation and capture that would derail progress. By contrast, Bangladesh and Zambia illustrated how growth strategies could be fashioned within the context of dysfunctional (and corrupt) governance. An important lesson Levy picked up from field work is that personalized competition, as in the latter two cases, can provide a workable platform for economic development.The book then tackles the principal challenge to development in many countries: the host government itself. He divides public sector reform efforts into hierarchical approaches (namely, strengthening the system-wide bureaucracy or programmatic coverage) and horizontal approaches (strengthening accountability mechanisms such as electoral systems and the judiciary). Since most countries offer limited opportunities for great leaps forward to good governance, the book explores options for easing governmental constraints at the program level that impede forward development momentum. That is, the development policy problem is disaggregated to programs and projects from the national policy level. For example, incremental and multi-stakeholder approaches for basic education in Zambia, and community-driven development programs in South Africa, each created “islands of effectiveness” in otherwise bleak public sector settings.In a nice dose of the reality faced by aid experts every day, Levy describes how in Africa as elsewhere public sector reform is complicated by the existence of “warring tribes” within donors and in country governments that, for instance, favor participatory approaches more than public management institutionalization. Professional tribalism is a well-known and serious problem at all levels of the development community, from NGOs to donors and private development firms, who are often split between bottom-up and top-down approaches. But the actual problem, as Levy notes, is how to develop the right balance of each for different contexts. As noted, democracy and good governance advocates often view their objectives as ends rather than means to development. Advocates have for decades tried to reverse-engineer the characteristics of affluent countries into shorter-term, low-income country reforms, ignoring the long historical struggles of advanced countries to make these very reforms. Levy calls this particular effort a “breathtaking combination of naiveté and amnesia.”Dividing up the problem, as Levy has done, can help reduce such tribal conflicts and allow forward movement. I learned from my own field experience in Latin America and Eastern Europe that while comprehensive public sector reform is practically impossible, incremental programs to build “islands of effectiveness” have stimulated reforms in ministries as stakeholder support builds up and “virtuous circles” multiply. Levy cites the example of the Bangladeshi garments industry, which now contributes 70 percent of total exports and employs more than 2 million people. Policy support for such “islands” required only targeted policy measures at the operational level to allow garments exporters to work around specific obstacles, such as financing inputs and meeting confirmed orders. They did not require broad-based policy reforms.I would also cite donor reforms of public financial management sub-areas in many poor countries at the operational level such as capital budgeting, public procurement, internal audit systems, and government financial management information systems (GFMIS) which have also supported the growth of virtuous circle efficiencies in particular sectors such as health and education. As an added touch of practicality, Levy offers many “public management lite” options for incremental reform gains in those countries where shifting to multi-stakeholder governance and accountability would be hard because of elite predation and patronage relationships, for example, building internal financial controls to control corruption and abuses of power, and initiating results-driven public management reforms in a few priority sectors as examples of what can be achieved.In all this meticulous analysis, however, one important question remains: What is the role of political culture in all this? Political culture behaviors—how people behave when outsiders are not looking—tend to be elastic and can serve as platforms for fundamental policy changes if approached tentatively, incrementally, and with the right incentive packages. They are the behavioral expression of linguistic, religious, and social values. The empirical problem with this concept is that studies have been unable to link particular types of political culture to specific political outcomes.Nevertheless, a while back my colleagues and I identified four types of historical and institutional contexts containing distinct “values and attitudes that affect political behavior”: Iberian, French, Former Soviet, and Great Britain and the United States. We found that these distinct cultural clusters affected the planning and implementation of fiscal decentralization programs.Motivation, political will, incentives, and timing are always thorns in the side of dynamic frameworks and typologies. So, one might ask of Levy’s frameworks, why in any type of regime would a “dominant leader” or “well-defined principal” necessarily emerge and persist? They may be only temporarily dominant, and then chewed up by local politics—making policy entries and sustained efforts impossible.Such behavioral and institutional variables point to the need to try to distinguish differences in political cultures as a means of deepening a development typology. For instance, Francis Fukuyama recently posited three dynamic development trajectories for Western countries similar Levy’s: the Anglo-American path; the Greek/Italian path, and the Prussian/German path. But these development journeys began differently and they were based partly on differences in political cultures. Thus, the Anglo-American path began with rule of law; the Greek/Italian with social mobilization; and the Prussian/German with building. Fukuyama’s trajectories pointed to the utility of political culture as at least a partial explanation for the success or failure of implemented development policies.Differences in political culture also explain why policy tools, frameworks, and lessons cannot simply be transferred at the strategic policy level from one country or region to another. Political cultures should be at the center of applied international comparative research. At the operational program level, for example, basic systems of public financial management, urban transport, health, and education, could be and have been adapted to obtain “good fits” in Levy’s terms. Such improvements at the operational level in the cases of Bangladesh and Zambia can be made by working with grain of their operational cultures. Political culture is more important if changes are expected at the strategic policy level, so it should be part of a practitioner’s tools to identify and measure constraints and opportunities for reform of key programs and sectors.Levy does cite Violence and Social Orders, in which Douglass North and his co-authors note that “culture” can determine whether transferred institutions and policies produce order or “unleash disorder” and make things worse. And elsewhere, he makes occasional mentions of culture, for example to partly explain Korean bargaining, equity, and Confucian role hierarchy. He also refers repeatedly to “existing interdependencies and networks” as the “grain” or foundation on which to build policies with which one must work overseas—implying political culture. But he does not formally employ the concept in his typology. With this minor caveat, it must be said this is a solid book which should be read by all development practitioners.
 George M. Guess, Jorge Martinez Vazquez, and William Loehr, “Fiscal Decentralization: A Methodology for Case Studies,” USAID Consulting Assistance for Economic Reform (CAER II) contract (Harvard Institute for International Development Discussion Paper #2, March, 1997).