Thursday, April 12, 2012

ADB Gives Warning On Rising Income Inequality

From the Beeb (excerpt):

Asia's rapid growth fuelling inequality, the ADB warns

Asia's rapid economic growth may undermine stability because the gap between the rich and poor is widening, the Asia Development Bank has warned.

Releasing its annual report, the bank said a key inequality measure increased to an average reading of 38 in Asia.

And while that is less than the average found in Latin America and Africa, Asia's figure is climbing as it declines in the other regions.

China, India and Indonesia have seen significant growth in inequality.

Speaking to the BBC, the Asia Development Bank's (ADB) chief economist Changyong Rhee explained that Asia may be seeing a long-term shift in the way the gap between rich and poor has been managed…

…However, over the past decade the sudden explosion of growth and rapid enrichment of many people has seen the rich-poor divide grow. The ADB estimates that currently in most Asian countries the wealthiest 5% of the population now account for 20% of total expenditure.

At the same time, for hundreds of millions of people access to education, healthcare and housing has become more difficult and expensive…

…In its report, the ADB said that the Gini coefficient in China had increased to 43 in 2010, from 32 in the early 1990s. For India, the figure rose to 37 from 33 during the same period. In Indonesia it jumped to 39 from 29…

…"Another 240 million people could have been lifted out of poverty over the past 20 years if inequality had remained stable instead of increasing, as it has since the 1990s," Mr Rhee added.

I can’t help but think that we’re only just seeing the countries singled out getting up to the Asian norm – looking at the NIEs, with the exception of Korea they all have Gini’s above 40.

Part of it is a consequence of rapid economic growth and changing economic structures. As economies shift towards higher productivity manufacturing and thence on to services, income disparities widen unless there’s significant counterbalancing fiscal intervention.

But that’s not to say that the situation is ideal or acceptable. It’s a balancing act – if growth is inclusive and benefits most of society through higher incomes and a higher standard of living, than widening income disparities are generally tolerated. It doesn’t matter if some benefit more than others, as long as everyone is dragged out of poverty at the same time.

But it’s when a certain standard of living is achieved that people start looking beyond the mere absolute level of income, and relative incomes become a more pressing point of discord.

Fairness in society is a nebulous concept, but one way to look at it is equal access to economic and social opportunities. If higher incomes leads to a concentration of asset ownership among the “rich”, that reduces economic opportunities for the rest. Not to mention that life outcomes depend to a certain extent on the level of income of your forebears. In that sense, it’s not “fair” – income and wealth inequality lead to unequal opportunities.

So rising inequality in those three countries are a concern, but the present level of inequality in the already middle-income and high-income countries in the region should concern us more.

Technical Notes:

  1. Original ADB press release
  2. "Asian Development Outlook 2012: Confronting Rising Inequality in Asia", Asian Development Bank, 2012


  1. There is an excellent article in Times (I think) very recently on why US economy recovery is yet perceived weaker than how it supposed to... Apparently since late 1970's US's wealth inequality has been in the rise.

    Though we can undoubtedly establish facts that it is perfectly normal for rising gini in face of economic growth, it however is politically and socially unhealthy. To top it off with ambiguous state governance we can't deny that it might evolve into something more critical for China.

    Suppose there is no effort to curb this unrest, it however will impact the economy... I think it safe to say though economically it is healthy (although not ideal) in long run wealth inequality might affect economy.

    But fact remains, inequality is not remote to Asian...

  2. Wazir al-Roslan al-ibnu KhaldunApril 12, 2012 at 8:46 PM

    Part 1

    Income inequality would remain an intractable problem as long as the a)core issues of access to capital and b)socio-economic policies that are skewed to favour the wealthy remain in place c) cultural practices are not reconfigured in light of changing times

    In the Chinese, Indian and Indonesian contexts, it would be safe to assume that much of that rise in inequality is attributable to differentials in capital ownership that can be leveraged upon to accumulate more wealth. I am talking of the landed class in the agricultural sector and the moneyed class in commerce who would have an inherent advantage over their peers in deriving more wealth. It is not surprising that income inequality has risen dramatically in the countries mentioned as government policies that favour manufacturing and service sectors meant an explosion of growth and consequently income levels in the urban centres of India, China and to a lesser extent Indonesia while the rural areas have been stifled by the low price strictures/subsidy structures that subsume the agricultural sector. And I have not mentioned about inputs for production like technolgy use in agriculture, small farm holdings, cultivation practices etc + the sudden explosion is also due to the absence of the "leveling" impacts of Socialism that strangled incomes in both China and India in different ways.

    If one moves across China from the coast to the interior, for instance, one would notice the urban-rural income disparities and the same within the respective intra regions between the moneyed/landed class and the proletariat/peasant classes. Add to that social outcasts like the itinerant workers who are denied citizen rights in the cities and we are probably looking at some of the major reasons. (an Indian equivalent would be the Dalits and the caste structure itself).

    It is interesting that Korea is the exception to the rule but I am not surprised. During a recent 1 week tour to Korea, I learnt from KDI officials of the extensive accessibility to education.(yeah, even monks in the interior are hooked on to Korea's fast broadbands to receive online tutoring for university degrees!)And because education is an empowering tool in fighting poverty, little wonder than they are rather immune to the problem. But then again, there are other factors at play including government policies, the collective rise of most Koreans from agrarian hell to industrial heaven etc. But then again, the Japanese have a relatively stable gap, but could that be also due the general stagnation Japan has been experiencing (the lost double decades?)apart from mass education, the "collective" industrialisation era, + pro-agrarian policies.

    Mind you, there are a lot factors involved for both income inequality and relative equality but space constraints dictate a general overview of the issue. Nevertheless I wish to share the PDF below which is rather informative about income inequality and its socio-economic implications:

    Warrior 231

  3. Wazir al-Roslan al-ibnu KhaldunApril 12, 2012 at 8:50 PM

    Part 2

    It goes without saying that income equality, particularly the nature of the gap is a combustible socio-political issue. Part of the recent problems in Malaysia is directly attributable to the increasing gap in incomes both inter-ethnic and intra-ethnic wise. While the former is somewhat acceptable to some (certainly not me though), intra-ethnic gaps are even more pernicious due to their insidious psycho-emotional effects like the "keeping up with the Joneses syndrome" particularly when purchasing power shrinks as wages fall behind the inflation curve.

    But few people realise that our neighbour down south is facing it even more acutely and that explains the decline in the popular vote and now a drastic call for re-engineering the model (which I reckon is "too little too late"):

    and the reaction:

    Interesting how the people there look at the issue in comparison to how we approach the minimum wage issue.

    Warrior 231

  4. Wazir al-Roslan al-ibnu KhaldunApril 12, 2012 at 10:37 PM

    Noticed my Part 1 is missing. Here is a re-post

    Part 1

    Income inequality would remain an intractable issue if core concerns such as a) accessibility to capital b) lopsided socio-economic policies c) cultural practices are not reconfigured in line with changing scenarios.

    If one analyses India, China and Indonesia, one is immediately struck by three nations emerging from the income leveling impacts of Gandhiist socialism, Maoism and Suharto statism. One would also observe how the reconfiguration of economic dynamics in each country has had a deleterious implications for income equality. Take China, for instance, its emphasis on SEZs, manufacturing, public construction and services sectors meant a transformation in the economy’s fulcrum away from the largely agrarian based / state industrial economy to a more capitalistic manufacturing based model. However, agrarian communities still remained as the provider of cheap food for the cities, the proverbial bread basket/granary for the towns .

    In addition, incentives devoted to the non-agrarian sector meant less investment in agriculture in terms of technology use, new cultivation practices etc that in turn led to a slower growth in productivity and hence incomes. This led to inter-regional income differentials with those in the hinterland badly hit. If one looks within regions themselves, the landed/moneyed class by virtue of their access to capital held an inherent advantage in the wealth generation stakes and that had a negative effect on income equality plus labour policies that depressed wages in order to attract investments contributed to a teeming low paid workforce.

    Thus, if one sweeps across China, the coastal regions have higher incomes than the agricultural interior and within these coastal regions, the moneyed class or connected party apparatchiks have more incomes than the working class proles. The same applies in the rural area where the landed class have higher incomes than the peasantry thus yielding both inter and intra-regional income differentials. Add to these core inequalities, such perverse practices like the deliberate ostracism of itinerant workers in Chinese society and you have the basis for income inequality. A similar scenario applies in India(with its caste systems and Dalits as added millstones) and Indonesia. And all three nations share another common feature; lack of accessibility to education and this where Korea comes in.

    I was on 1 week trip to Korea recently and had the opportunity of discussing with some KDI friends. Believe it or not Korean’s accessibility to education is phenomenal (yeah, even monks in the interior could access the knowledge portals of their cyber universities via their super-fast broadbands!!). As we know education is the slayer of poverty. Little wonder, Korea does not have a Gini problem like India, China or Indonesia. Yet another reason lies in history: Korea “collectively” industrialized transforming itself from a largely rural agrarian backwater to a humming industrial giant and in the process income levels rose equitably like riding on a rising tide. It was a case of everybody poor becoming everybody wealthy or thereabouts over same time frame. And in the process of industrializing, they dint neglect agriculture, incentivizing it and letting market forces decide prices in most cases, hence rural income levels also rose there you have it in a nutshell why income inequality is not that apparent in Korea. Ditto too Japan which shares many similarities.

    Space constraints dictate a general overview of factors but i would like to share this PDF with readers of this blog as it provides insights into income inequality and its implication in Asia:

    Warrior 231

  5. @Abbas

    No, Asia's not the only, or even the worse, in terms of inequality. But as you say, that's not an acceptable excuse. Bear in mind though that Malaysia (and Singapore's and Hong Kong's) Gini numbers are worse than the US, and above the global average.

    @warrior 231,

    The income gap between races has actually been closing, but the gap within each community has barely budged and is about the same across all. It's a Malaysian problem, not a Malay or Chinese or Indian problem.

    Korea's and Japan's low Gini numbers are partly due to cultural factors, but also due to active government intervention (transfers from rich to poor), something which the rest of the region simply doesn't do at all.

    Thanks for the links. You might also be interested in these:

  6. Wazir al-Roslan al-ibnu KhaldunApril 13, 2012 at 8:05 PM

    Part 2

    Without being irascibly argumentative, allow me, an engineer not an economist by training, to put forth additional observations on the issue:

    1. the simplistic premise that it is time to jettison the NEP by dismantling the Pribumi affirmative frameworks flounders in the quagmire of its own overinflated contention that the NEP has been the source of the inequities plaguing the Malaysian socio-economic landscape. In fact, a close perusal of the data would indicate the Gini was at 0.44 in 1990 (the last year of the NEP) while at 2009 ,19 years after its gradual dismantling, [via Brother Anwar bin Ibrahim's (BABI for short)pro-market policies and Dol Badawi and Najib's monkey-handed reconfigurations of the policy] the Gini was still mired at 0.44!!!

    2. There is a mistaken assumption amongst many experts that the "level playing" field of unfettered capitalism is the panacea for narrowing inter-ethnic income gaps when nothing can be further than the truth. In the scenario envisaged above, the Chingks with their vast hoards of fixed and liquid assets will emerge the winners in "a fittest will survive survivor series" and there is no prizes for guessing why this would the inevitable outcome.And mind you, capital is only part of the tale of the tape as clannish wholesaler, middleman, retail networks and financial frameworks come into play.

    3. The notion that hard work and the provision of equal opportunities offers a rational solution to the issue is a purely naive assumption (2) for no level playing field exists in the first place and simply because hard work can be trumped by street smarts and inherited wealth anytime!For that matter, anyone care to spin me another grandmother yarn of Malay rags to riches, Chingk style (for that was a myth all along)

    For Malaysia to erase the inequality curse there is only one palatable solution, a wealth tax targeting the wealthy and a one off re-compensation tax levied on the Chingks, This latter tax would be reminiscent of this:

    "Putin’s one-time windfall tax appears to be based on the U.K. approach, says Morgan Stanley. In 1997, the U.K. imposed a one-time payment on the owners of assets acquired during privatizations in the 1980s. The tax was 23 percent of the difference between the purchase price and the average net profit during the first four years following privatization, multiplied by the factor of 9.3. This resulted in an extra $8 billion for the British crown, according to Morgan Stanley”

    Imagine how such a "windfall" + a wealth tax would go a long way towards re-engineering the Pribumi socio-economic landscape and erasing the effects of years of exploitation the Chingks foisted upon us, the Pribumis of this beloved land.


    1.BABI's minimalist and naive 'blame the gov' take on the issue when he was primarily responsible Post 1990:

    2. (maybe the gentleman here is of the same type I bump into often during my semester breaks in California all those years ago. (Melayu Mudah Lupa akan Kaumnya)

    3. (your take last year)

    Extra links of Interest



    Warrior 231

  7. @warrior 231,

    (You got hit by Blogger's spam filter again)

    With reference to the first of your latest posts.

    1. The gini coefficient within races (like that of Malaysia as a whole) has not changed much at all in the last 25 years. In reference to your contention that Chinese incomes have been "steaming ahead", try calculating the ratios, and then the growth rates of incomes. As of 2009, Bumi household incomes on average are about 90% of the Malaysian average and 72% of the Chinese average, up from 79% and 56% in 1992. Virtually all of those gains came in the last ten years.

    (I'm cheating, I've got a much longer series than the link you provide, from 1970 up to and including the 2009 data).

    The biggest difference between the NEP years and the post-NEP years is that the "quick wins" had been exhausted - most of the reduction in inequality came from attacking the problem of poverty and from the gradual replacement of British expats with local talent i.e. the ends of the income distribution. It might interest you to know that the ratio of Bumi and non-Bumi corporate holdings both rose during the NEP years.

    Is the income difference between Bumis and non-Bumis a factor in our present state? Yes it is and will continue to be for some time (the rate of convergence suggests something like 20-30 years). But it's increasingly overshadowed by income differences between top and bottom, irrespective of race. If anything, income differences between the top and bottom for Bumis is somewhat worse than the rest.

    The fact that the overall gini, and gini within races, has not changed despite these Bumi income gains points to something much more complex and fundamental than the "simplistic" story of poor Bumis and rich Chinese.

    This isn't an argument against affirmative action, but rather a more nuanced approach to it.

    2. Wealth inequality is almost always higher than income inequality. The reason why it's not commonly used as a measure of inequality is for the same reason that wealth taxes aren't common - valuation difficulties. Wealth inequality is also a function
    of income and inheritance. Higher income taxes, and especially implementing a capital gains tax should account for the former, while estate/inheritance taxes can address the latter.

    3. Problem: Much of (rural) Bumi asset ownership is tied up in property. But because much of it is Malay reserve, which artificially depresses property values, little of it came be leveraged by Bumi owners as collateral or liquid capital. That's an enormous handicap. This is of course a constitutional issue, and unlikely to be resolved soon, if at all.

  8. In reply to part 2:

    1. (*cough*) The Bumi to Non-bumi income ratio actually fell slightly during the 1990s. So no argument here, except to note that the gini within races also stayed pretty much the same. That suggests that the NEP policies are not specifically to blame. The flip side to that of course is that strengthening blanket pro-Bumi affirmative action policies is unlikely to help either.

    2. I agree. Surprised? Don't be - I'm not a fundamentalist market advocate. Wealth generates wealth.

    3. A wealth tax is difficult to implement without being unjust. That's why I think an estate/inheritance tax would be fairer and more doable, since it directly addresses inter-generational transfer of wealth without reducing the incentive to work and invest, while reducing the potential for resistance since its a one-time deal for each generation.

    Second, it should be implemented across all races, not just on the Chinese.

    Third, we also urgently need a capital gains tax, to reduce gains from asset shuffling and to equalise the playing field between long term investors and speculators.

    Fourth, the tax yield from these taxes should be directed at lower income groups and rural population (especially in East Malaysia), not for other purposes. Rich urban Malays shouldn't benefit.

    Fifth, subsidies need to be rationalised, because they disproportionately benefit higher income households. Whatever savings again should be directed where its most needed.

  9. Dear Hishamh,
    you said the income difference between Bumis and non-Bumis will converge something like in 20-30 years.

    At present rate and economic structure..I seriously doubt it.

    Seems even credible economist like Hishamh have been misled by the reducing C/B and I/B income ratios.

    Its clearer if we look at the income graphs - the chinese/indian income curve's trajectory is way way steeper than the Bumi's (almost flat) the twain will never meet. The curves are diverging - not converging, sir.

    These income ratios will reduce but only as fast as the exponential decay(half life) of Lynas' radioactive waste!

    So contrary to what you suggest - in 20 or 30 years the gap will be much, much bigger.

    But that's quite logical - no mystery!

    Unlike the EPPs, the diasdvantage rakyat and the Bumis are not getting enablers, when they do - they're mere peanuts.

    Like you said, wealth generate wealth,..and poverty reproduce poverty.

    What you suggested as measures to address inequality are already widely known and discussed: capital gains tax, inheritance tax, targeted subsidies, etc..and even tax on wealth.

    What boggles the mind - if they are obvious solutions..why hasn't the economists in the country especially in PEMANDU, EPU and ISIS forwarded such measures to the Government or even attempt to push for them?

  10. Dear Hishamh,
    What is inclusiveness and sustainability as mentioned in the NEM and ETP?

    For that matter, what also is High Income?

    Does inclusiveness means 80% of the Malaysian population will earn hand-to-mouth peanut incomes with zero disposable/ discretionary balance?

    Does sustainability refers to social sustainability or environmental sustainability? If 80% of the rakyat earns peanuts, will the country achieve political sustainability?

    And c'mon, does High Income of US15k GNI per capita means 80% of the rakyat earning peanuts?

    What @##@%&!&^% going on?

  11. @anon 2.01

    When I was studying stats, one of the first things to do was to plot the data. I didn't come to this conclusion purely from looking at the ratios - those only served to confirm what my eyeballs are telling me.

    Not sure what data you're looking at. From 1987 to about 2002, the income lines diverge slightly. But since then the Bumi income line is steeper than the others. Mathematically, its hard to have an increasing ratio unless the raw data is actually converging or parallel.

    Some of the sample data (from 1987 to 2002 which was the most recent lowest point for the ratio and gap, to 2009):

    1987 averages: Bumi - RM868, Chinese - RM1488 (gap RM620, ratio 58.3%)

    2002 averages: Bumi - RM2376, Chinese - RM4279 (gap RM1903, ratio 55.5%)

    2009 averages: Bumi - RM3624, Chinese - RM5011 (gap RM1387, ratio 72.3%)

    Increase in incomes since 1987 (i.e. the slope of the income lines): Bumi - 317.5%, Chinese - 226.7%

    Increase in incomes since 2002: Bumi - 52.5%, Chinese - 17.1%

    So not only has the ratio of Bumi incomes to Chinese incomes increased, the actual gap in Ringgit has also been closing since 2002 - I don't see how you can interpret this in any other way.

    As far as these measures being proposed to the government, its been done but there isn't much public pressure on the government to change the tax structure, mainly because these will be deeply unpopular particularly with urban voters.


    High income status, taken on its own, means squat. I think I've said that before :)

  12. Wazir al-Roslan al-ibnu KhaldunApril 18, 2012 at 1:14 AM

    Part 1
    1.The divergence in our perspectives with regard to inter-ethnic inequality basically stems from the approaches adopted. If a quantum approach is adopted, then the data will show that Malay cohort is actually lagging behind the Chinese cohort. If we define the data in percentiles, of course there would be the illusion of the Malay cohort gaining when in fact the Chinese are starting from a higher baseline denominator. Allow me to illustrate what I mean (Focus is on top 20 percent using my data until 2007).

    2. If we analyse it in terms of raw figures between 1992 and 2007 , the data will depict a gain of approximately RM 4566 for the Malays and a RM6530 increment for the Chinese. However, if we take the “median” 1999 data into account, a clearer picture emerges wherein the quantum gains are 1755 for Malays and 3122 for the Chinese respectively which in real terms means the Chinese gained by an additional RM637 (after the relevant subtractions) over the Malays between 1999 to 2007. And if we spread the respective ethnic gains over the 15 period between 1992-2007, per annum calculations would divulge that the Malay cohort averaged gains of 4566/15 = 304.40 per annum versus 6530/15 = 435.33 for the Chinese.

    3. If we take the percentage gain perspective, then we will obtain a % gain of 165% for the Malays in contrast to a 122% gain for the Chinese over the 1992-2007 period but then again the Chinese baseline denominator would be about RM 2248 higher than the Malay baseline to begin with. Of course by utilising the ratio measure, a perceptible narrowing in income inequality would be apparent.Whatever the perspectives adopted, any gains, narrowing, divergence or convergence between the Malay and Chinese derived should be offset against the larger demographic reality wherein the higher fertility rate of the Malays over the Chinese (almost double) would by default mean a significant drag on Malay per capita income data due higher dependency ratios and household sizes. An additional point to note,is the fact that the highest surge in income between 1980-2008 involved the top 20% cohort in which the Chinese dominate as illustrated by figure 19 here (1).

    Warrior 231

  13. Wazir al-Roslan al-ibnu KhaldunApril 18, 2012 at 1:27 AM

    Part 2

    4. The second misapprehension of the argument relate to the NEP. The suggested nuanced approach instead of the outright jettisoning of the NEP is noted but the recommendation that it should shift its orientation towards East Malaysia and rural Malaysia at the expense of Bumis in urban Malaysia is misplaced. If we look at public expenditure patterns during the immediate aftermath of 1990 particularly during BABI’s reign, you will note a diminution in public expenditure contributed to a spike in the Gini with the Malay cohort most affected by BABI’s (makes me wonder some Malays fall under the spell of this Chinese lackey) pro-market paradigm. In actual fact, if you look at Tables 4 and Tables 5 here (2), when public development expenditure was reduced in the 1991-1995 period, the Malay/Chinese ratios increased from 1.74 to a high of 1.83 in 1997 at the tail-end of BABI’s tenure implying clearly that any reduction in such expenditure will have deleterious consequences for the Malay cohort. In fact, the urban/rural ratios spiked to a high of 2.04 in 1997 from 1.70 in 1990. To be precise, the targeting of expenditure patterns do have an impact on income inequality which suggest that the nuances you recommend should be focused on recalibrating the targeting component. For instance, there should be shift away from public expenditure in wholesale and retail trade which benefits the Chinese cohort to expenditure on education which benefits the Malay cohort (page 151 of (2)

    5. Despite my disagreement over his ethnic-neutral postulations in addressing income inequality, Roslan’s (2000) perceptive paper (3) on the issue is revealing as to the extent of Anwar bin Ibrahim’s decimation of income inequality among the ethnic groups and particularly within the Malay cohort. This lends credence to the suspicion that the said economic cleavages created in a hitherto economically homogenous Malay community were primarily designed for political reasons and these fissures were deliberately allowed to widen so that marginalized segments within the community could be exploited for political benefit. The collective angst of this exploited segment was eventually cross-yoked with marginalized segments of other ethnics in order to form a broad-based urban-rural proletariat/peasant corp to be manipulated by a Chinese/Christian dominated elite and their pseudo-Malay mandore lackeys. The discourse of marginalization and oppression by the capitalist class is now a common theme in the PR narrative and Anwar is on record as depicting the upcoming polls as a framework for class warfare.

    In fact, the historicity of both PAS, DAP are deeply rooted in leftist /communist ideology, if one dissects their respective epistemology whilst Anwar’s tutelage under such socialist or left-leaning luminaries like Dr Syed Husin of Parti Rakyat, Kassim Ahmad, Rustam Sani is well-known..... .but then I diverge...........

    Warrior 231

  14. @warrior 231

    Yes, thank you for an interesting, and civil, discussion.

    However, Wazir, there are a couple of problems with using quantums instead of ratios, and with your analysis. With a ratio, you're comparing Ringgit to Ringgit at one point in time. With quantums, you're comparing Ringgit to Ringgit across time, and that's problematical because a Ringgit in 2007 doesn't have the same purchasing power as a Ringgit in 1992.

    With inflation, the Ringgit gap would appear to grow larger even if purchasing power doesn't change, whereas the ratios would stay precisely the same (yes, I checked). To get a more accurate measure of the quantum gap, you have to adjust for inflation, thus allowing for a true Ringgit to Ringgit comparison.

    Secondly, 1999 isn't a good year to choose as your "median" because it was a recovery from recession year - incomes were highly depressed. Top 20% bumi incomes dropped by RM340 from 1997, while top Chinese incomes dropped by about RM770.

    Third is that there is a structural break in the data - I pointed it out in my earlier comment that Bumi incomes lost some ground against Chinese incomes until from 1987 to 2002, and then gained rapidly afterward. That's very clear from the ratio plot, but it's not something you can see using the raw data. Taking averages across the 1992-2007 sample period isn't meaningful, because you're not capturing the dynamics of what's actually going on.

    Fourth, the Gini coefficient is most sensitive to changes in the centre of the distribution of incomes, so its probably best to use the mid-40% households rather than the top-20%.
    To be honest though, taking the middle cohort doesn't make a difference to the general trend.

    So if you're going to use quantums, I'd suggest breaking up the sample in to two periods (1987-2002 and 2002-2009), and adjust for inflation.

    You can get the relevant income data here:

    For the 1987-2002 period (still using top 20% incomes for comparisons sake and converting all incomes to 2000 prices), Bumi incomes generally lost ground against Chinese incomes.

    The ratio dropped from 58.3% to 53.6%, and the income gap widened from RM1,890 to RM4,905. Chinese incomes gained RM6,032 compared to Bumi RM3,018 over the same period. Compound annual growth rates were 5.8% p.a. for Chinese incomes and 5.2% for Bumi incomes.

    I have no problem with agreeing that Bumi incomes gains did indeed lag Chinese incomes during this period.

    But from 2002-2009, we have a radically different picture. Again adjusting for inflation, the ratio improved from 55.5% to 72.3% and the income gap narrowed from RM4,905 to RM2,607.

    Bumi incomes increased by RM1,705, while Chinese incomes fell (that's right, they dropped) by RM592. Compound annual growth rates were 3.8% p.a. for Bumi incomes, while for Chinese incomes it was -0.8% p.a.

    If you plot the graphs of Chinese incomes adjusted for inflation from 2002 onwards, middle income Chinese households have seen their incomes flatline (0.3% p.a. CAGR), while incomes of the top 20% have dropped. Low income Chinese households still saw income growth but it was below average (1.1% p.a. CAGR).

    No wonder nobody supports MCA anymore.

    In case anyone's wondering, the same thing is going on with Indian incomes - and that explains MIC's predicament.

    Whereas Bumi incomes have not stopped rising across all income segments, though strongest for the top 20% (3.8% top, 3.6% mid, 3.5% low CAGR 2002-2009).

    The next HIES will be conducted either this year or next. It will be interesting to see if these trends continue. If my hypothesis is correct (and I need to test it out first), then we should still see Bumi income gains outstrip Chinese income gains for the next survey (2012-2013), but with a possible reversal over the next two survey periods (2015-2019), before a resumption after 2020.