Malaysia’s growth is leaving cities behind

Many countries run their urban economies the same way: investment and opportunity concentrate in a handful of large cities, while mid-sized cities with real productive capacity contribute well below their potential. The result is a system under strain at the top and underperforming everywhere else. A new World Bank report on unlocking state-level growth in Malaysia documents this story with evidence. Using two diagnostic tools — Zipf’s Law and an Economic Potential Index (EPI) — the report identifies specific mid-sized cities as places with real strategic assets that are not yet converting them into growth. Zipf’s law: Reading an Urban System's Health Zipf's Law offers a simple way to assess whether a country's urban system is functioning as it should. In a balanced system, city sizes decline predictably by rank: the largest city is roughly twice the size of the second largest, three times the size of the third, and so on. When cities deviate from this pattern, it signals that population and investment are concentrating disproportionately in a few large centers, putting strain on infrastructure, pushing up living costs, and drawing resources away from the rest of the system. Malaysia's cities are moving further from this benchmark. Between 2010 and 2020, inequality in city sizes increased, driven by the rapid expansion of already-large urban areas. The country's biggest cities kept growing; the rest of the system fell further behind. Figure 1. Large cities grew larger between 2010 and 2020 From Diagnosis to Action: the Economic Potential Index Knowing that the urban system is out of balance is useful, but it does not tell policymakers where to act. That is where the EPI comes in. It combines factors such as human capital, market access, and agglomeration economies to estimate what each district should be producing given its resources. Comparing that potential against actual economic performance produces a simple and useful classification: districts that are overperforming relative to their assets, and districts that are underperforming, where policy attention is most likely to pay off. The highest-priority group is districts with high potential but low performance. In Malaysia, these districts collectively generate 28 percent of total national district GDP, a significant share produced below capacity. They tend to have real strategic advantages: skilled workforce, existing infrastructure, or strong market access. What they lack are the conditions to convert those assets into output, whether due to regulatory constraints, weak connectivity, or the absence of targeted investment. Figure 2. Mid-Sized Cities: High Potential, Low Performance When the Zipf and EPI findings are read together, a specific set of cities stands out: mid-sized, high-potential, and underperforming. They are large enough to have real productive capacity but not so large that growth there would add to the congestion already straining Malaysia's largest centers. The investment data reinforces the problem. Between 2011 and 2021, Malaysia's already high-performing states, Selangor, Kuala Lumpur, and Penang, recorded the highest investment growth rates. Lagging states, Terengganu, Perlis, and Sabah, recorded the lowest. Investment followed performance rather than potential. A Different Policy Approach The implication is not that investment in large cities should stop. It is that the allocation logic needs to change. Right now, investment tends to follow performance, flowing to places that are already doing well. The Zipf and EPI framework points toward a different principle: direct resources toward places where the gap between potential and performance is largest and where returns to new investment are likely to be highest. What Governments in Other Developing Countries Can Do The underlying method is replicable. For governments in other developing countries facing similar patterns of geographic concentration, the evidence points to a set of concrete actions. One way to unlock this potential in the short term is through systematic peer learning. Cities and municipalities can share experiences and adopt proven solutions from high-performing urban centers. By adapting such best practices to local contexts, mid-sized cities can overcome inefficiencies without reinventing the wheel. This peer-to-peer approach not only accelerates reform but also fosters collaboration. Over the longer term, empowering mid-size cities through policies that prioritize education, healthcare and digital connectivity will improve residents’ quality of life and ease migration pressures on major hubs. Strengthening local infrastructure and services in these cities would enhance the city’s livability, expanding the local labor market, which in turn makes the cities attractive for workers and business alike, opening up new economic opportunities and ultimately becoming new centers of growth to address regional economic disparity. In addition, establishing regional partnerships and bolstering local governance capacity will help these cities manage growth effectively and sustainably. A Lesson Beyond Malaysia Any country experiencing geographic imbalances in growth can apply a similar framework: use a Zipf diagnostic to assess whether the urban system is concentrating too fast, then use a potential-performance index to identify where the underperformance is sharpest and where the case for intervention is strongest. The value of this approach is that it moves the policy conversation from generalized concerns about regional inequality to specific, evidence-based decisions about where public and private resources are most likely to generate returns. Acknowledgement This blog post draws on a working paper reviewed by Dr. Kwok Tong Soo, whose comments as external peer reviewer materially improved the underlying analysis.