Wednesday, December 1, 2010

2009 State GDP

Just a quick note – I’m a little under the weather from a cold, so I’m taking it easy for the next few days.

This report was actually released last week, and almost brings us up to date on state level GDP. If you recall, DOS started releasing state GDP data about this time last year, starting with 2005-2006 data. The latest numbers report 2009 GDP figures by state.

I haven’t gone so far as to analyse any of the numbers (the October MSB release has priority), but there don’t appear to be too many surprises here – Penang was the worst hit by the recession, due to its high exposure to export manufacturing, while KL continued growing because of its high concentration of service industries.

Technical Notes:

GDP Report By State from the Department of Statistics

7 comments:

  1. Hishamh,
    Is there a co-relation between GDP per capita (or GNI per capita) and annualized "mean household income" per capita. Any Idea?

    Any data or research done on this that u know of?

    Wouldn't that relationship be relevant in respect of the NEM/ETP's goal of achieving higher income status for malaysia by way of increased GNI per capita.

    ReplyDelete
  2. Alan, as I understand it, they use household income as an input into estimating GDP - so the measures are related.

    But bear in mind that household income is determined by an irregular survey (currently 2-3 years), whereas GDP is determined quarterly (and in some cases in much higher frequencies e.g. weekly and monthly). The methodology for national income accounts is better established and easier to compare internationally.

    ReplyDelete
  3. There is a dearth for household income data from sources available on the net but I have the 1999 household incomes by state.

    Basically I wanted to compare household income and GDP per capita.

    The GDP per capita by state from the url you provided is only are for year 2006 upwards so I chose the data for 2006 to compare.

    Okay there is a 7 years gap there but I just want to see the trend.

    I had the mean monthly household incomes annualized, then convert them to per capita using household size of 5.0.

    Big gaps for KL and Penang:
    WP KL GDP per capita: 44,341
    Hsehold Inc/capita /year: 9,852
    Ratio: 4.5

    Penang GDP per capita: 30,490
    Hsehld Inc/capita /year: 7,507
    Ratio: 4.1

    Interestingly, the gaps get smaller in the poorer states:

    Sabah GDP per capita: 10,921
    Hsehld Inc/cap /year: 4,572
    Ratio: 2.4

    Kelantan GDP per capita: 6,012
    Hsehld Inc/cap /year: 3,154
    Ratio: 1.9

    Kedah GDP per capita: 10,386
    Hsehld Inc/cap /year: 3,869
    Ratio: 2.7

    Care to comment?

    ReplyDelete
  4. Alan, here's some of the adjustments you have to make:

    i) changes in inventories
    ii) less corporate savings
    iii)add imports, less exports
    iv)add taxation, and less govt spending

    it's not quite a one-to-one conversion.

    You should also consider that household membership is highly heterogeneous between urban and rural areas, and can change radically over time. GDP/GNI per capita has somewhat the same type of bias due to evolving dependency ratios, but these don't change quite as drastically as household memberships can.

    ReplyDelete
  5. Sorry, I should add that differences in household memberships probably explain most of the differences you've calculated, especially between rich states and poorer ones - poorer households tend to also have more members.

    ReplyDelete
  6. I agree household size has bearing. But that would be too easy.

    What's worrisome is the emphasis on Greater KL.

    The ETP planners had obviously ignored the already glaring disparity betwen the rich states and poor states, which if one looks carefully highlights high congruency between economic status of the states with ethnicity.

    Putting some RM 50 billion (a large chunk of the govt funding for ETP !) into greater KL will not really benefit the greater population. Multi-plier or multi-spanner, oh really !?

    I also doubt it'll significantly improve the houshold incomes of the people in Greater KL themselves - having observed how Gamuda operated in their previous mega projects – especially when they’re all underground.

    Okay, the 'abstract GNI per capita numbers' will somewhat improved, no doubt. What would Idris Jala's ETP strategy be if his KPIs are household income (esp of the poorer communities), GINI, HDI, etc.
    But Idris Jala though savvy in management know-how (at least know how to spend govt's money on expensive foreign management consultants) is "so old school" in national economic development and sustenability.

    Like I heard from Chomsky and modified:

    The poor always complain,
    They always do,
    But that’s just idle chatter,
    Our ETP brings rewards to all,
    at least to all that matter.

    Does the poor in Kelantan and Sabah matter?

    ReplyDelete
  7. Alan, I don't disagree with any of what you're sayng - it's just that household income isn't a terribly good altenative measure.

    I've never felt the NEM was strictly speaking necessary for the achievement of high income status, and more importantly that high income status was a goal worth achieving in itself. There will be losers in this development process, and there doesn't appear to be much in the way of remedial action to limit the impact on those left behind. Many of the ETP projects appear designed to raise corporate incomes, not individual incomes, which will exacerbate income inequality.

    But the issue with household income is that to be useful it has to be comparable between states at any point in time, and across time - that isn't the case.

    To give an example, my grandfather's family household numbered 10, my father's 7, and the average among my siblings and I is a little under 3. In my experiemce that seems typical across generations, and the movement between rural to urban households. On that basis, household income isn't a good measure.

    GNI per capita of course says nothing about who's actually getting the income either, so it has problems too. I'd actually want to look at real wages per worker, but that data is only collected for the manufacturing sector.

    ReplyDelete