This is a joint post with Julia Clark and Christian Meyer.
Industrial policy—as many have already commented—is back. (See here, here and here).
The recent wave of post-financial-crisis interventionism has reignited the classic (and often heated) debate about whether governments can in fact nurture economic growth. Previous analysis of the East Asian miracle, and frustration at the perceived failure of certain liberalization policies, has led many to (again) embrace a more activist role for governments in economic development.
Chief among them is World Bank Chief Economist Justin Yifu Lin, whose “New Structural Economics” (NSE) framework combines a market-friendly economic system with proactive industrial policy centered on a country’s comparative advantages. The theory, detailed in Lin’s new book, is the subject of much debate: is NSE a road to successful industrialization, or will its attempt to “pick winners” be a slippery slope to greater government failures?
Last week, CGD hosted a discussion to unpack these questions with Lin, moderator and Senior Fellow Alan Gelb, and panelists Anne Krueger, John Page, and Steven Radelet. The lively discussion that followed Lin’s presentation provided a few answers, raised many questions, and outlined key issues for thinking about industrial development.
What’s new about New Structural Economics?
Building on previous work, the NSE approach emphasizes the importance of the structure of a country’s factor endowments (read: the composition of its labor force, natural resources, human and physical capital) in determining opportunities for industrialization and the appropriate role of the state. Given this structure, each country has comparative advantages in certain industries. Lin argues that, in the past, industrial policy failed when countries adopted strategies that "defied" comparative advantages and thus required copious amounts of protectionism to survive (if they did at all).
Each stage along the development continuum (from Australia to Zambia) implies a different structure, which in turn requires different infrastructure. The dynamic process of moving from one development stage to another thus requires the evolution of this infrastructure (both hard, like transport and telecom; and soft, like financial regulations and worker training). While markets are central to allocating resources efficiently at each stage, investors in poor countries, especially the “first movers”, face many coordination problems and externalities. Lin argues that states must also play a role by facilitating infrastructure upgrades. In addition to providing general infrastructure, NSE advocates targeted government support of specific industries in line with the country’s comparative advantages, as suggested by exports of somewhat richer countries and the existing domestic industry. This idea of “picking winners” makes NSE a relatively activist—and controversial—variant of industrial policy.
Old challenges for NSE
Understanding the role of structural change in economic development remains important for many developing countries, and NSE is an important contribution to this effort. However, despite Lin’s emphasis on the “new” in NSE, the panel raised some old but important critiques of industrial policy, that any country (or donor) seeking to promote industrial growth would be wise to consider:
- Picking winners isn’t easy: In order for the NSE approach to work, governments must be able to first identify their comparative advantages, and then decide what appropriate sector to promote. Knowing that a country with a comparable abundance of unskilled labor has successfully developed a textile industry doesn’t necessarily help you decide whether to promote textiles in general or to specialize in synthetic fiber or nylon. These are not easy tasks, even for seasoned industrialists.
- Officials may not be up to the task: Compounding the general difficulty of picking winners is the fact that this approach relies on bureaucrats to do so, which pose additional problems. Unless officials have a high level of business acumen—or sufficiently embed themselves in the business sector—they may lack the necessary information to make wise choices. Selection may be compromised by officials prone to cronyism and corruption. This is particularly true for the poorest countries, which may have the biggest market failures but also the weakest government capacity to intervene.
Beyond industrial policy
Acknowledging both the merits and challenges of the NSE approach, perhaps it should be seen as only one component in an already existing—and much milder—arsenal of public policies, such as trade promotion, industrial agglomerations, FDI strategies, and programs that attract and build firm capabilities. The panel discussed this broadly and hit on a few key points, including:
- Firm capabilities: The skills and working practices of firms are a key driver of productivity and the basis of any firm’s competitive advantage. In context of the wider discussion about industrial development, many discussants pointed to an apparent lack of firm capabilities, especially in African manufacturing firms. Even if the government targets the “correct” industries for growth, firms may not be competitive if they lack technical capacity. One reason for the scarcity of larger-scale and longer-term capability-building programs was seen in the pressures and incentives in the aid industry, which tends to prefer projects with fast and tangible results.
- Spatial industrial policy: Beyond country-wide reforms and industrial policies, the panelists also discussed smaller-scale interventions targeted at creating specific, sub-national enclaves in which firms can then operate under improved conditions. Examples for such agglomerations include export processing zones, industrial parks, or free trade zones. If successful, enclaves can subsequently be integrated into wider industrial strategies.
In the end, many questions as well as many puzzles remain. Can governments really tailor the NSE model to fit their diverse economies and histories? Would they be able to avoid the pitfalls of previous industrial policies? How might Lin’s theory correct or address the processes of de-industrialization that have dogged many African countries in the past? How can we incentivize long-term investments in human capital, firm capabilities and government capacity in an era of results-based aid?
To understand these issues better, we need careful empirical research that monitors efforts to accelerate structural upgrading and competitiveness. In the end, results will likely answer these questions better than ideology.