Understanding and Unlocking State-Level Growth Potential in Malaysia

Over the past two decades, richer states have outpaced poorer states in growth. All states saw an increase in income, as indicated by the ratio of 2023 per capita GDP to 2005 per capita GDP being greater than one for all states. Furthermore, a positive correlation exists between per capita GDP growth rates from 2006 to 2023 and initial per capita GDP in 2005.

As richer states have grown faster than poorer ones, income inequality between regions has widened, resulting in divergence. The divergence trend in Malaysia contrasts with the convergence observed in China, Brazil, and the European Union.

States in Malaysia have accelerated their potential output growth, in contrast to the global trend of slowing potential growth. Collectively, Malaysia's states saw potential output growth average an estimated 4.4 percent per year from 2001 to 2010, rising to 5.1 percent per year between 2011 and 2021. This growth trend contrasts sharply with the global trend among EMDEs, where potential growth fell from 6 percent to 4.9 percent.

Based on potential output growth estimates, poorer regions in Malaysia had the opportunity to converge and catch up to the leading states. In both decades, potential growth estimates were higher for lower-income regions, allowing for the conditions of convergence. If the states had grown at their potential growth rates, convergence could have been achieved.

Most of Malaysia's states have underperformed relative to their potential output growth, with actual growth falling short of potential in recent years. The difference between actual and potential growth is larger for lower-income regions compared to higher-income states.

The absence of convergence in Malaysia suggests that the assumptions of the neoclassical model do not fully hold. Barriers to factor mobility and the agglomeration of certain sectors, among other factors, may account for the increasing disparity.

Internal migration is low and declining (0.46% in 2022), non-pecuniary factors like cultural norms and family ties keep people attached to local areas, land ownership rules in Kelantan restrict capital mobility, and poorer states have limited structural transformation from agriculture to services/industry.

How is the Economic Potential Index different from potential growth? The previous key messages examined potential growth, which estimates how fast a state's economy can grow based on its available inputs: workers, capital, and productivity. It is a measure of speed. The Economic Potential Index (EPI) takes a different approach. It asks what building blocks a district has on the ground, such as infrastructure, connectivity, education levels, market access, and institutional quality, to assess how well-positioned that district is to support economic activity. It is a measure of foundation. A district can score high on the EPI (good infrastructure, educated workforce) but still grow slowly if there are frictions preventing it from realising that potential. The two measures complement each other: potential growth tells you the ceiling, while the EPI tells you how solid the foundation is.

The EPI reveals significant regional disparities at the district level. States like Kuala Lumpur and Labuan top the EPI scores, while poorer states such as Sarawak, Sabah, and Kelantan show lower average EPI scores. However, there is notable variation within states.

Having measured each district's economic potential through the EPI, the next step is to assess whether districts are actually living up to that potential. The EPI provides an estimate of the economic output a district should be able to achieve, given its available resources (infrastructure, human capital, market access, and institutional quality). Using nighttime lights data as a proxy, the report estimates each district's actual GDP per capita and compares it against the predicted GDP per capita from the EPI model. A district whose actual output exceeds its predicted level is classified as overperforming; one that falls short is classified as underperforming.

This comparison produces a two-by-two framework that classifies districts along two dimensions: economic potential and economic performance. The vertical axis captures performance (whether a district is producing above or below what its endowments predict), and the horizontal axis captures potential (the EPI score). Together, they sort districts into four quadrants, each with distinct policy implications.

Evidence of spatial spillovers in Malaysia reveals that districts near high economic activity hubs exhibit stronger economic performance. NTL growth in Peninsular Malaysia is primarily driven by peripheral areas near urban centers, reflecting a pattern of urban sprawl. Established urban centres like Kuala Lumpur, Johor Bahru, and George Town experience relatively lower nighttime light growth, while their adjacent areas grow faster, as surrounding suburban and peripheral zones catch up with fresh investments, new housing, and industrial projects.

The urban sprawl phenomenon is further reflected in the increasingly unequal distribution of city sizes across Malaysia. Zipf's law suggests that a country's largest city should be about twice as large as the second-largest, three times as large as the third-largest, and so on. In Malaysia, city sizes are less equally distributed than predicted by Zipf's law, and the inequality in city sizes has increased between 2010 and 2020. Bigger cities have grown disproportionately faster than smaller ones, driving Malaysia's city size distribution further away from what Zipf's law would predict.

Malaysia is a highly centralized federation, with the federal government exercising powers and responsibilities that typically belong to subnational governments. According to the Constitution, the exclusive powers of the federal government extend to areas such as education (encompassing elementary, secondary, and tertiary levels, as well as vocational training and teacher regulation), healthcare (including hospitals and clinics), and water supply. These areas are traditionally assigned to subnational governments in most other federations. As a result, state governments are left with limited functional responsibilities, primarily land, agriculture, forestry, and local governance.

Since states have a smaller functional role, it follows that their revenues and expenditures are lower compared to subnational governments in other federations and even some unitary systems. Malaysian state revenues accounted for only about 21 percent of total general government revenues, while state expenditures accounted for only about 12 percent. In most federal countries and in several unitary systems like Japan, Korea, Indonesia, and Vietnam, subnational governments are responsible for over 30 percent of public spending.

Each state government's ability to generate its own revenue and its reliance on federal transfers varies significantly. Between 2018 and 2022, federal transfers accounted for less than 30 percent of total revenue in Selangor, Sabah, and Sarawak. Conversely, federal transfers comprised more than 70 percent of total revenue in Perlis and Terengganu. For the remaining states, these transfers constitute approximately 50 to 60 percent of their total revenue.

There seems to be no clear set of rules or formulas guiding the allocation of certain federal grants. As of 2022, the federal government disburses at least 15 types of grants to state governments. Among these, only three have their allocation rules explicitly provided in the Constitution. Capitation Grants and State Road Maintenance Grants constitute about two-thirds of total transfers. The allocation criteria for the remaining grants are generally outlined without explicit rules, leaving significant discretion to the executive branch.