In much of the developing world, economic power is largely concentrated in the hands of a “market-dominant” ethnic minority. The classic case is southeast Asia, where the Chinese, usually a tiny proportion of the population, enjoy an overwhelmingly dominant economic position. In Malaysia, the average Chinese household had 1.9 times as much wealth as the Bumiputera (Khalid 2007); in the Philippines, the Chinese account for 1 per cent of the population and well over half the wealth (Chua 2003). The same is true in varying degrees in Indonesia, Burma, Thailand, Laos, and Vietnam. Southeast Asia is an acute but by no means isolated example, from Jewish oligarchs in post- Soviet Russia, who are six of the seven richest men in the country; to Indians in east Africa; to the Lebanese in Sierra Leone and the Ibos in Nigeria, the picture that emerges is that certain groups of ethnic minorities often do disproportionately well in economic terms (see Chua 2003). Naturally, disparity between the economic power of an ethnic minority and the disadvantaged position of the majority ethnic group is a potential source of political instability. However, market-dominant minorities have received surprisingly little attention from economists.

In a recent paper, we provide a valuable piece to the literature by taking Malaysia as an example and analysing the evolution of income inequality among different ethnic groups. Most importantly, as the racial issue has long been a fixture in Malaysian politics, by studying the impact of Malaysia’s long-standing affirmative action policies on income distribution, we hope to provide insights on economic policy design for mitigating ethnic tensions and improving social cohesion in Malaysia and other countries facing the same issue.

From the 1970s to 1997, the growth rate of the per adult real national income in Malaysia has been 2.93 per cent. After a short setback following the Asian financial crisis, the economy has been catching up with an even stronger trend, and from 2002 to 2016, the growth rate of the per-adult national income was 3.7 per cent. However, the benefits of growth always remain a contentious issue, especially in a plural society such as Malaysia, with its multiracial, multireligious and multi-ethnic population.The largest ethnic group in the country is Bumiputera, a Malaysian term describing Malays and other indigenous peoples of Southeast Asia — it literally translates as son of the soil. In 2016, the population consists of approximately 68 per cent Bumiputera, 24 per cent Chinese, 7 per cent Indian, and 1 per cent others.

Inherited as a legacy of British colonial policies, the Bumiputera have remained the poorest group with the lowest average income, compared to the relatively richer minority contingent of ethnic Chinese and Indians, since Malaysia gained independence from in 1957. This economic imbalance, especially along racial lines, was a recipe for disaster. It naturally led to political and social instability, and racial riots erupted slightly more than a decade after the country gained its independence. As a response to the race riot in 1969, the government developed a comprehensive affirmative action plan known as the New Economic Policy (NEP) in 1971.

The NEP was formulated with the overriding objective of attaining national unity and fostering nation-building through the two-pronged strategy of eradicating poverty and restructuring society. Especially, while the first prong is class-based, the second prong was designed to elevate the socioeconomic conditions of the Malays. The strategy was not new; in fact, the NEP expanded the affirmative action as enshrined in the Constitution. Article 153 of the Constitution specifically highlighted the special position of the Malays and this inclusive growth policy continued to be adopted throughout Malaysia’s post-NEP economic history and was included in the National Development Policy (NDP) (1990-2000) and National Vision Policy (2000-2010). The inclusiveness agenda continues in the New Economic Model (2010-2020), where the policy goal is for Malaysia to become a high-income country by 2020 as well as sustainable and inclusive; the latter is defined as “enabling all communities to fully benefit from the wealth of the country” (National Economic Advisory Council, 2010).

The evaluation of NEP and its successors is inconclusive. On one side, to some extent, it achieved remarkable results by reducing poverty from nearly 50 per cent in 1970 to less than 1 per cent in 2014. The income gap also shrank. Household income inequality, as measured by the Gini coefficient, narrowed from 0.513 in 1970 to 0.446 in 1989 to 0.410 in 2014. On the other side, schemes favouring Malays were once deemed essential to improve the lot of Malaysia’s least wealthy racial group; these days they are widely thought to help mostly the well-off within that group, while failing the poor and aggravating ethnic tensions. Interestingly, the introduction of the NEP was a direct result of the racial clash triggered by severe losses of the ruling government in the 1969 general election, where they lost more than half of the majority votes. In a recent election (2018), as a dramatic twist, the coalition that ruled Malaysia uninterruptedly since its independence lost to an opposition coalition. The root cause was not unlike the cause of political changes in the past — the Malays felt that the benefits from growth did not trickle down to them, and only the well-connected groups (read: cronies) who were involved in corruption enjoyed the fruits of development. The recent election also showed that the majority of Chinese voted for the opposition, as occurred in 1969.

By combining information obtained from national accounts, survey data, and fiscal data, this paper attempts to construct the first Distributional National Account (DINA) for Malaysia and systemically address the fundamental question mentioned above: In terms of income growth, which income class and ethnic group benefits from economic growth and to what extent, especially considering that Malaysia has an extensive race-based affirmative action policy? To obtain a full picture of the evolution of inequality in Malaysia, ideally, we would carry out the analysis for the period 1957-2016, but due to data limitations, we focus on the period 2002-2014, after the Asian financial crisis (AFC). In other words, we provide only partial answers to the question posed earlier: Who benefited from post-AFC economic growth?

A. Evolution of income inequality

Figure 1. Income shares: top 1 per cent Malaysia vs. top 1 per cent in other countries (pre-tax national income)

Notes: Distribution of pretax national income (before all taxes and transfers, except pensions and unemployment insurance) among adults. Equal-split-adults series (income of married couples divided by two).). Imputed rent is included in pre-tax fiscal income and pre-tax national income series.
Figure 2. Income shares: top 10 per cent Malaysia vs. top 10 per cent other countries (pre-tax national income)

Notes: Distribution of pretax national income (before all taxes and transfers, except pensions and unemployment insurance) among adults. Equal-split-adults series (income of married couples divided by two).). Imputed rent is included in pre-tax fiscal income and pre-tax national income series.

Figures 1 and Figure 2 compare our Malaysian DINA top income series with the DINA series recently computed for the US (Piketty, Saez and Zucman, 2018), France (Garbinti, Goupille and Piketty, 2018) and China (Piketty, Yang and Zucman, 2019). These series use the same methodology as the one applied in this paper: they all attempt to combine information obtained from national accounts, surveys, and fiscal data to estimate the distribution of pretax national income (including undistributed profits and other tax-exempt capital income) among equal-split adults.

In 2002, Malaysia’s inequality level was extremely high: its top 1 per cent income share was 19 per cent and the corresponding number for the top 10 per cent was 44 per cent, which is higher than those of the US and substantially higher than those of China. However, while we observe a trend of increasing inequality after 2002 in the US and China, Malaysia’s inequality has been decreasing. By 2014, we see that income inequality in Malaysia is much lower than it is in the US and similar to the level in China, but still significantly higher than the level in France.

Table 1. Income growth and inequality 2002-2014: Malaysia vs. other countries

Sources: China: this paper. China: Piketty, Yang, Zucman (2019). USA: Piketty, Saez and Zucman (2018). France: Garbinti, Goupille-Lebret and Piketty (2018). Distribution of pre-tax national income among equal-split adults. The unit is the adult individual (20-year-old and over; income of married couples is split in two). Fractiles are defined relative to the total number of adult individuals in the population. Estimates are obtained by combining survey, fiscal, wealth and national accounts data.

Table 1 compares the distribution of 2002–2014 real income growth in Malaysia, the US, China and France. Aggregate growth has obviously been different in the four countries. As emerging economies, both Malaysia and China have experienced exceptional growth, especially China. The average per adult national income has increased by 55 per cent in Malaysia (corresponding to an average annual increase of 3.7 per cent) and almost tripled in China, while it has increased by 9 per cent in the US and 3 per cent in France for the same period.

Despite the significant economic upswing, Malaysia’s growth is featured by its strong inclusiveness. From 2002 to 2014, the growth rate accruing to the bottom 50 per cent has been significantly larger than the growth rate of the top 10 per cent (which is much larger than that accruing to the top 1 per cent). This is in stark contrast with the growth rate in China and the US. Especially, in the US, for the same period, total growth accruing to the bottom 50 per cent is -7 per cent compared to 29 per cent to the top 1 per cent. The result for France is similar to the result for Malaysia; e.g., the income per adult in the bottom 50 per cent is growing faster than that of the top (the top 1 per cent and the top 10 per cent). However, the average real income per adult growth rate is much lower in France than in Malaysia for the study period; e.g., the total cumulated growth is 0.2 per cent in France and 3.7 per cent in Malaysia.

B. Evolution of income inequality by ethnicity group

In this section, we begin by estimating the share of the population that are Bumiputera, Chinese and Indians for different income groups, especially for the top 1 per cent and the bottom 50 per cent Malaysian adults. This will allow us to respond to important questions regarding racial disparity in Malaysia.

Figure 3 illustrates the population share of Bumiputera, Chinese, Indians, and other ethnic groups by percentiles of real income per adult for 2002. It is quite striking to see how the share of the Chinese increases when approaching the top, contrasting with the sharp decrease in the share of the Bumiputera. Of the total Malaysian adults in 2002, 61 per cent are Bumiputera, 30 per cent are Chinese, and 8 per cent are Indians; however, in the top 1 per cent, Bumiputera account for only 24 per cent, Indians account for 3 per cent and the Chinese account for 72 per cent. Clearly compared to Bumiputera and Indians, Chinese are over-represented. This gap was mitigated in 2014; however, the contrasting pattern persists: among the richest one percent of Malays. (See Figure 4).

Figure 3. Population shares of ethnic groups in each percentile, 2002 (pre-tax national income)

Note: Population shares are calculated for each 2 percentiles using Kernel-weighted local polynomial smoothing method.
Figure 4. Population share by ethnic group – the top 1 per cent income group (pre-tax national income)

As important as it is to understand the unbalanced population distribution at the top, it is equally crucial to look into the poorest segment (e.g., the bottom 50 per cent Malaysian adults), especially when policymakers and the general public evaluate affirmative policies. As Figure 5 shows, in 2002, in the bottom 50 per cent of Malaysian adults, 73 per cent were Bumiputera, 17 per cent were Chinese, and 9 per cent were Indians. Until 2014, the trends of the three ethnic groups were very stable. Thus, when pro-Bumiputera policies improved the economic status of low income Bumiputera, approximately one-quarter of the population in the bottom 50 per cent was left behind. This segment of the population is made up of non-Bumiputera Malaysians, approximately 29 per cent of total Chinese and 56 per cent total Indians.

Figure 5. Population shares of ethnic groups in the bottom 50 per cent (pre-tax national income)

We now further analyse and decompose the income share (e.g., the top 1 per cent, the top 10 per cent, the middle 40 per cent and the bottom 50 per cent) by ethnic groups. As shown in Figure 6, the decline in the top 1 per cent share is dominated by two trends: a strong decrease in the share of Chinese and a significant increase in the share of Bumiputera. Among the top 1 per cent of Malaysian adults, the income share of the Chinese decreased by almost half, from 15 per cent in 2002 to 8 per cent in 2013. The share of the Bumiputera doubled in the same period, increasing from 3 per cent to 6 per cent. The share of the Indians also increased significantly and more than doubled from 0.3 per cent to 0.7 (however, in terms of the absolute level, the effect is minimal). The results for the top 10 per cent of Malaysian adults are similar to those for the top 1 per cent.

Figure 6. Decomposition of top 1 per cent income shares by ethnic group (pre-tax national income)

Figure 7. Decomposition of bottom 50 per cent income shares by ethnic group (pre-tax national income)

Using the same procedure, we decompose the bottom 50 per cent income shares. The results for the bottom 50 per cent are different from those of the top 1 per cent and top 10 per cent. The substantial expansion in the bottom 50 per cent income share was solely driven by the increase in the share of Bumiputera, e.g., from 11 per cent in 2002 to 14 per cent 2014, while the share of Chinese and Indians stagnated at 3 per cent and 2 per cent, respectively (see Figure 7). For the middle 40 per cent, the moderate increase of the income share can be decomposed to a steady increase in the share of Bumiputera (from 20 per cent in 2002 to 24 per cent in 2014) and a slight decrease in the share of Chinese (from 16 per cent in 2002 to 15 per cent in 2014).

In conclusion, the decrease in income inequality in Malaysia was mainly driven by two opposite trends: a sharp decrease in Chinese income shares in the top and a substantial increase in Bumiputera income shares in the top and bottom.

By further decomposing the pretax personal income share for both Chinese and Bumiputera in the top 1 per cent by income sources, (e.g., wage income, self-employed income, property income, and transfer income), we found, strikingly, the decrease in the share of the Chinese at the top is mainly driven by the decline of the Chinese property income share, from 9 per cent of total pretax personal income in 2002 to 3 per cent in 2014. Meanwhile, the increase in the share of the Bumiputera is driven almost equally by the increase in wage income, self-employed income and property income (the income shares for each type of income source increased by 1 per cent from 2002 to 2014). We conduct the same exercise with the share of the Bumiputera in the bottom 50, and the results show that approximately two-thirds of the increase in the Bumiputera share is driven by the increase in the Bumiputera wage income share.

Finally, to address the core question of this paper: In terms of income growth, which income class and ethnic group benefits from economic growth and to what extent, we decompose the average growth rate of real per adult national income in Malaysia by both income groups and ethnic groups (Figure 8). Although Malaysia’s growth for the period is featured by its strong inclusiveness, macro growth has obviously been different in the three ethnic groups. In particular, in the top 10 per cent, the average growth rate per adult national income for Bumiputera is 5.4 per cent, compared to 1.2 per cent for Chinese and 4.6 per cent for Indians. The gap in the growth rate among the different ethnic groups was large in the top 10 per cent; however, it is even larger for the top 1 per cent. In the top 1 per cent, the average growth rate for Bumiputera was 8.3 per cent, which is sharp contrast to -0.5 per cent for Chinese and 3.4 per cent for Indians.

Figure 8. Decomposition of growth rate of real income per adult, 2002 to 2014 (pre-tax national income)

Compared to the top income group, the difference in the average growth rate among the ethnic groups are not significant for the middle 40 per cent and bottom 50 per cent. For the middle 40 per cent, the average growth rate for all three ethnic groups was 4.1 per cent, and for the bottom 50 per cent, the Bumiputera’s growth rate of average per adult national income was 5.4 per cent, which was slightly higher than that of the Chinese at 4.9 per cent and Indians at 4.7 per cent.

The most important implication of Figure 8 is that although the middle 40 per cent and the bottom 50 per cent benefited significantly from economic growth, the Bumiputera in the top income groups (the top 1 per cent and the 10 per cent) benefited the most from economic growth. In sharp contrast, the income of the Chinese in the top income groups deteriorated. In a way, the strong growth in high-income Bumiputera occurred at the cost of a decrease in Chinese and the slow growth of Indians in the top income groups.

To conclude the findings of this paper, Malaysia’s growth for the period of 2002-2014 included significant and relatively egalitarian (among different ethnic groups) growth in the middle 40 per cent and bottom 50 per cent and a strong divergence among the Bumiputera, Chinese and Indians in the top 1 per cent and top 10 per cent. Thereby, Malaysia’s growth features both an inclusive redistribution between income classes as well as a combination of affirmative action policy and the free flow of capital(ism).

♣♣♣

Notes:


Muhammed Abdul Khalid is the economic adviser to the Prime Minister of Malaysia. Prior to this appointment, he led the Council of Eminent Persons’ (CEP) secretariat. The council advises the government on economic and financial issues. He has served as a consultant for the World Bank, the United Nations Development Program (UNDP), the United Nations Children’s Fund, and the United Nations Economic and Social Commission for Asia and the Pacific. Muhammed obtained his PhD from the Institut d’études Politiques de Paris, France, and holds a Master’s degree in economics from the University of Malaya and a Bachelor of Science degree from University of Southern California, Los Angeles.

Li Yang is a postdoctoral researcher at the Paris School of Economics (PSE) and coordinator for East and Southeast Asia at the World Inequality Lab (WIL). He’s interested in wealth and income inequality, economic development and economic history. Email: li.yang@psemail.eu