Thursday, January 16, 2014

Potential Growth In ASEAN

From the latest round of IMF working papers (abstract):

Potential Growth in Emerging Asia
Rahul Anand ; Kevin C. Cheng ; Sidra Rehman ; Longmei Zhang

Summary: Using three distinct approaches—statistical filtering, production function, and multivariate model— this paper estimates potential growth for China, India, and five ASEAN countries (Indonesia, Malaysia, the Philippines, Thailand, and Vietnam) during 1993–2013. The main findings include: (i) both China and India have recently exhibited a slowdown in potential growth, largely reflecting a decline of total factor productivity (TFP) growth; (ii) by contrast, trend growth for the five ASEAN countries has been rather stable and might even have increased marginally, with the notable exception of Vietnam;(iii) over the longer term, demographic factors will be much more supportive in India and some ASEAN economies than in China, where working-age population should start shrinking, with the overall dependency ratio climbing by the end of this decade. Improving or sustaining potential growth calls for broad structural reforms.

As the abstract says, the general consensus around the various approaches used suggests that potential growth has slowed from declining TFP growth in China and India, while being mostly stable in ASEAN.

For Malaysia, potential growth has dropped slightly from the pre-2008 period, also largely due to declining TFP growth, though this decline is fairly slight – eyeballing the chart, it looks like somewhere between 0.5% to 1%. Right now, potential (non-inflationary) growth is in the region of 5.0% or a little below, which agrees with my own primitive and rudimentary estimate.

While the estimation of potential growth is methodologically sound – at least, I have nothing to criticise – the analysis is somewhat less so. Take the graphs on Pg. 12, from which the authors derive some conclusions about the evolution in potential growth. The first three charts show purported relationships between R&D expenditure, infrastructure development and economic complexity against TFP growth, but:

  1. China is such an obvious outlier in all three charts, it should have been taken out before estimating any relationship; and
  2. But after taking out China, the correlations become really weak, and you’re deriving conclusions from a sample of only 6.

The fourth chart doesn’t try to show a correlation, but leads to a real howler of a mistake in the text (Pg. 9, emphasis added):

Nevertheless, trend TFP growth remains typically low in these five economies, particularly compared to China, and also, to a lesser extent, India. This could reflect a host of factors, ranging from: low Research and Development (R&D) expenditure (particularly Indonesia, the Philippines, Vietnam, and Thailand), poor infrastructure (particularly Indonesia and Thailand), low levels of economic complexity (particularly Vietnam, Indonesia, and the Philippines), and difficulty in doing business and stringent regulations in product markets (particularly Malaysia and Thailand) (Figures 7–10).

In the World Bank’s Ease of Doing Business Indicator, a low number indicates a high ranking. Far from being hard to do business in, Malaysia and Thailand are the highest ranked countries in the sample, and the easiest to do business in. Both countries are in fact ranked among the top 10% of the countries in the world (12th and 18th easiest respectively).

So take the numbers with some confidence, but you might want to take the conclusions and policy advice with some healthy scepticism.

Technical Notes

Rahul Anand & Kevin C. Cheng, Sidra Rehman, and Longmei Zhang, "Potential Growth in Emerging Asia", International Monetary Fund Working Paper No. 14/2, January 2014


  1. Hello Hisham

    1) does this study dispel myth that Malaysia is losing out on our SEA neighbours?

    2) apologize for being out of topic

    PwC did this study on impact to cost of living due to GST implementation at 4%

    since the current announcement of 6%, has any similar study been done?

    1. Dear Hisham

      Another totally off topic question. again much apology.

      I wanna ask your opinion on this

      1) in 1991, Canada introduced a Federal level GST at 7% replacing a 13.5% Manufacturers' Sales Tax.
      2) I tracked Consumer Price Index and compared to US (which is using SST) and found CPI increased throughout 1991 then reduced again which is consistent with a one time increase in prices (eventho GST rate is lower than the Manufacturing Sales Tax)

      3) After that, i noted in August 1990 there was an oil shock due to the 1st Gulf War which caused crude oil to increase from USD17/barrel to USD30+/barrel. the oil shock lasted for 9 months

      4) So I tracked Core Consumer Price Index, which removed energy and food cost, hoping to get a more accurate view price movements. found that from 1991, Core Consumer Prices in Canada decreased vis a vis the US.|%20%20CANADA%20CORE%20CONSUMER%20PRICES

      5) so does this suggest decrease in tax rate was passed to consumers in cheaper price? but CPI itself doesnt reflect that due to the 90s oil shock

      would love to hear your views on this. shooting in the dark here.

    2. @anon

      To your first set of questions:

      1. I don't understand - since when were we losing out to our neighbours in terms of growth?

      2. Treasury has the data, but its not published...yet.

      For the second set of questions, you can refer to a research paper I linked to.

  2. 1) Here are some articles leaving me that impression. Among my friends, it is a prevailing view too.

    2) oh nice. is it good or bad news?

    I calculated, based on the PwC data. GST at 6% will cause CPI to decrease by 0.07%. hope it is in that region. at least CPI dont increase. (0.1/(10 - 6) * (10 - 4) = 0.07

    3) given that is a state level experience, how much does it relate to a national level implementation?

    I noticed the implicit elasticity for the Canadian provinces is quite high (0.6 for 'All items' category). based on the PwC data, ours for CPI is a mere 0.02 (0.1/(10 - 4) = 0.02). does that imply malaysia has an inelastic demand?

    Is there any national level implementation that mirrors our situation?

    when I think about it, we are quite unique. We have a Sales Tax and Service Tax regime and we are switching to GST at a lower rate. almost everybody that I know switched from just a sales tax on goods (eg Canada's Manufacturer's Sales Tax; Australian's Wholesale Sales Tax;UK's Purchase Tax) to GST. The taxing of services (which makes big chunk of GDP share) is surely gonna lead to increase in price. meaning most countries' GST experience is not the best reference for us

    1. @anon

      That's an awful lot of poor analysis and misinformation to correct.

      So dealing from top to bottom:

      1. Koon Yew Yin's entire thesis is not supported by the data. For reading on the subject of corruption and growth, I wrote a whole series of posts two years ago:

      Also, the idea that we should have a higher level of development because we are resource rich is in complete defiance of global economic history:

      And Malaysia has performed just as well or better than Korea and Taiwan over the last decade.

      2. For the contention by Anwar and others regarding falling behind Singapore and Indonesia, this ignores the different structures of the economies in question. Both Malaysia and Singapore are trade dependent; Indonesia is not. Hence Indonesia's growth has barely been affected by the global crisis. Bear in mind however that Indonesia is starting from a very low base, and will have a naturally higher growth rate, as is discussed in the paper I linked to in this blog post.

      The difference between Singapore and Malaysia mainly relate to different export mixes, with Singapore concentrated in pharma, oil re-exports and E&E, while Malaysia is both commodities based and E&E. To a very great degree, Malaysia's economic growth and trade performance is tied to global commodity prices, which is not something that either the government or the opposition have much control over.

      I should also note that the difference in the size of the Malaysian and Singapore economies reached a low point in 1998, when SG's economy was 18% larger. Malaysia has since overtaken SG again, and should now be 8%-9% larger, based on IMF data.


    2. [cont]

      With respect to GST, Treasury indicates that there will be an increase in the CPI by 1.9% for GST at 6%. The reason for the increase is mainly due to the much larger coverage relative to the sales and service tax. Some prices should come down (furniture; cars), but many items will be taxed for the first time.

      I should note however, that not all the services sector will fall under GST - government services, finance and insurance services are exempt from GST. Also, some segments really collect GST on goods, not services (e.g. wholesale and retail trade), and hence already part of the SST framework.

    3. 1) Thank you for the answer and links. It is surely a different narrative from the 'we screwed up' one bandied around so freely. malaysia being a commodity based nation in the past is at the mercy of volatility and crashes.

      that explains the graph below. coinciding with tin price crash, Korea went over the hills and far away from us

      2) the question is does that mean resource rich nations can obtain better results by private companies to fully invest and develop their natural resources? the nations just collect dividends from profit sharing. and work on developing other industries from the proceeds without spending a cent.

      3) btw, what is E&E?

      4) On GST, so it means the new tax will contribute 1.9% to 2015's CPI, correct? that doesn't count price increase from other sources right? eg energy cost, labour cost etc. so we prolly gonna see full CPI of 4% in 2015?

      5) I am curious. on a percentage basis.. how many items are taxed and how many are exempted under SST?

      from the source below, it was said 3k items are exempted under SST. I guesstimate there are shud be at least 10k items in the market, so most items are already taxed. Apparently that's not the case

      6) btw what can consumers do? based on my understanding of tax incidence, a more elastic demand will see businesses taking a bigger hit on tax burden and passing more of tax reduction to consumers. seems like we as consumers could do somethings to improve our chances.

    4. @anon

      1. Yes, the prices of tin, palm oil, crude oil, all more or less crashed during the course of the 1980s. Kind of hard to keep up when your export income halves.

      2. Whether natural resources are managed publicly or privately is almost irrelevant. Whether KLK or Felda are producing palm oil, doesn't change prices on global markets.

      3. Electrical & electronics

      4. Correct, though it will take some time for the price increases/decreases to go through. There's going to be a settling in period for businesses to manage the transition, which could take a few years.

      5. Of the items included in the CPI, only about 45% are subject to SST. This will increase to about 75% under GST.

      6. Businesses incur no tax under GST, unlike under SST. Under GST, businesses only act as collection agents. However from the consumers perspective, its a lot more transparent under GST.

    5. 2) but doesnt it mean govt resources is freed up to invest in other industries not intertwined with commodity (eg E&E factories). thus when commodity crashes, we are less hurt by it

      3) thanks! i am a n00b

      4) you mean that 1.8% increase can be ammortized? eg 0.6 in 2015, 0.6 in 2016, 0.6 in 2017

      5) there are about 1k items inside the CPI basket of goods right? outside the CPI basket, can number of items seeing price decrease offset number of items seeing price increase in CPI basket of goods?

      6) if in the past seller sells at RM10. now the price is RM10.60 after tax. is it possible in an elastic market, seller considers reducing price to remain competitive -- perhaps RM10.30 after tax. so tax burden is shared among consumers and businesses. how do consumers get that result?

    6. @anon

      2) The government shouldn't be investing in factories. Governments are terrible at running businesses. What Malaysia has done is to encourage investment in non-commodity based industries, after trying the other way with Proton, Hicom et al.

      In any case, you're missing the point - a commodity crash directly impacts GDP irrespective of whether those commodities are produced by the public or private sector.

      3) You're welcome

      4) No, it's not an ammortisation process. What's likely to happen is that only the bigger companies will be hooked into the GST system at first, before spreading out into the rest of the economy. There's an annual revenue floor of RM500k for collecting GST, and some might be hesitant to declare they make that much revenue in a year.

      5) Short answer - no.

      6) In practice, some items will have sales tax embedded in the price, so you might not see the price go up so much. Secondly, all prices must be quoted without GST, so I doubt there will be any cuts i.e. the price will stay quoted at RM10, with GST charged on payment.

    7. 1) I note that commodity crash will affect GDP irregardless of who's producing. But a diversitied economy would be able to withstand commodity crashes better (eg 2009 recession). so, take from 60s till late 70s, tin and oil prices were on the rise: if M'sia let private companies handle those industries (while still receiving dividends) and channel all its money to encouraging investment (not run business, my bad!) in non-commodity industries, then we might be more diversified to tidal through the crash in the 80s. that's what im thinking.

      4) i see. wouldn't that mean that if there were more early adopters, CPI increase would have been lower because more business can make use of input tax credit

      5) were you surprised by this? given your analysis in GST and inflation II (reduce in rate is passed to consumers in cheaper or worst case scenario -- same price). It is quite surprising to me cause Malaysia already impose sales tax on good and service tax. I thought our items coverage were already pretty wide. apparently not.

      6) price not going up so much? meaning before GST or after GST?

      based on GST Customs, price will be quoted inclusive of GST for consumers

    8. @anon

      1) Yes, you're correct. Unfortunately, hindsight is 20/20.

      Also, we should consider that the industrialisation phase that Malaysia went through in the late 1980s and 1990s was a fortunate confluence of events, coinciding Malaysia's desire to enter light manufacturing and Japan's desire for a lower cost manufacturing base. Those conditions did not exist in the 1960s and 1970s.

      4) That's possible, but frankly we're all speculating here.

      5) Yes I was. I didn't realise how narrow the coverage was for SST. For example, something like half of the revenue from the sales tax comes from just one source - cars.

      6) After GST.

      If prices are going to be quoted inclusive of GST, that's against global best practice. Somebody is not thinking right.

    9. 1) Agreed. it is easy to comment with hindsight. but on privatizing natural resources and government focusing on growing non commodity industries -- has it been done before in the developing world? I can't think of any developing states that is not heavily involved in running its O&G sector for exp. the benefit is there (eg higher probability of a diversified economy). perhaps the political and security concern of privatizing your resources make it untenable -- esp given bad experience during colonization

      5) is there any plans (or proposal from the finance industry) to channel extra revenue from GST to social assistance for the poor and middle class? 1.8% increase in CPI is pretty frothy. if govt cut subsidy later this year or in 2015 -- 2015 might push people over the edge. even now the chorus of complains is deafening. najin is even talking about intervention

      6) wouldnt be the first time people at the top are not thinking right

      7) the other day. i floated an idea that malaysian businesses are profiteering. and u compared our GOS being higher than Singapore. I understand GOS is gross output minus price intermediary goods and employees compensation. but how does that relate with profiteering? is it that the 5-10% extra are businesses not passing savings from subsidy and price control to the consumers?

      btw. thanks for taking time to comment. always appreciated

    10. @anon

      1. Actually, prior to 1975 most of Malaysia's O&G industry was private-sector. Petronas was only set up in late 1974, and didn't become fully integrated (exploration, production, refining) into the industry until the 1980s (full disclosure: my father joined in 1978, and spent the rest of his career there). Tin, which was a big export item before the 1980s, was also mostly private sector managed. Palm oil was a bit of both, with Felda being counterbalanced by mainly British owned and run plantations. The palm oil GLCs we see today were mostly taken over in the early to mid 1980s (all credit to Khalid Ibrahim for that).

      5. As part of the GST rollout, there will be a one-off increase in BR1M, plus the income tax cuts. That's it as far as I know. Also note: there will be no significant increase in revenue for at least 3-4 years after GST implementation.

      7. Flip your question on its head. You can deal with inflation by looking at prices, or you can approach the problem from the other direction by looking at incomes. If incomes are growing as fast or faster than price increases, than cost of living increases are moot (from a consumer perspective).

      The fact that Malaysia's GOS is relatively high says that companies have not been sharing their excess profits with workers. Technically, the income share of capital is too high, that of labour too low; alternatively, you could say that the margin of profit is too high.

      Either way, managing cost of living increases can be handled by increasing the share of national income going to labour. BR1M is one manifestation of that effort.

    11. Hishamh,

      GOS is relatively high .. esp. the share of capital is too high compare to that of labour. What would be the right ratio?

      Zuo De

    12. @Zuo De,

      I'm thinking we should aim for 40% over the long term.

    13. Is this the ratio found in developed nations? Or something that you feel socially? Rather, what is the basis to arrive at 40%?

      Zuo De

    14. @Zun De

      1) to my knowledge GOS = Gross Output - Compensation for Employees - Cost of Intermediary goods & services. I found this data: Compensation of Employees by GDP for the EU. (I'm guessing) this is the Compensation of Employees portion of the calculation.

      2) Employees Compensation by GDP (%)

      EU Countries = 49.9% of GDP
      Denmark - 55.0%
      Sweden - 54.2%
      Norway - 45.6%

      3) Compensation of employees alone take almost 50% of gross output. Minus 10-20% for Cost of Intermediary Goods & Services, we would come to the 40% target hisham mentioned

      4) I wonder is there a correlation between Compensation of Employees with Income equality -- Sweden, Norway and Denmark have high percentage of income share and at the same time they are the best scorers for income equality. But have to be noted UK's Compensation of employees is over 50% as well but they don't fair as well income equality (overpaid footballers?)

    15. @Zuo De

      It's more of an intermediate target, which I think we can get to. Most developed economies are around 50% or higher. But that might be a bit of a stretch.


      Great leg work!

      But while your definition of GOS is correct, its not quite accurate to put it that way in terms of national income.

      That would be:

      GDP = GOS + CE + Mixed income

      There's a couple of other terms related to depreciation and taxes/subsidies.

      So what I'm looking at is GOS at 50%, COE at 40% and Mixed income at 10%.

      On point 4, what you're looking for is tax rates. The UK has probably the lowest marginal tax rate and the lowest share of government in the economy among the major EU states. Also its not footballers really, but overpaid bankers.

    16. Just a quick rundown on the current status of employee compensation, based on the latest available data, which unfortunately dates back to 2007-2010. All data is a percentage of output (sales):

      Construction: 21.7%
      Manufacturing: 5.4%

      And in services:

      Lawyers: 32.7%
      Accountants: 44.9%
      Architects: 26.9%
      Engineers: 20.3%
      Surveyors: 29.0%
      Private schools: 34.4%
      Medical services: 21.0%
      Dental services: 21.8%
      Private hospitals: 18.5%
      Accommodation: 19.7%
      Brokers: 16.2%
      Real Estate agents: 25.9%
      Advertising: 11.9%
      Cinemas: 6.3%
      Bus transport: 22.8%
      Road haulage: 13.2%
      Shipping: 7.9%
      Air transport: 8.0%
      Trains: 27.4%
      Highways: 3.3%
      Telecoms: 5.5%
      Computers: 17.1%
      Wholesale: 2.2%
      Retail 4.8%

      Sad, isn't it? To be fair, some of these sectors are highly capital intensive (e.g. telecoms), but when even lawyers are getting shafted...

    17. Thank you all, this is really great help for me, guide me in my business.

      Ya, I thought the footballers were to be blame but indeed the Bankers / Wealth management characters. They are terrible like that guy in Singapore about poor man transport (public transport) vs his Porsche.

      Zuo De

    18. @hisham

      so, the equations goes..
      1) GOS = Gross Sales - Compensation of Employees - Cost of Intermediary Goods and Services

      2) in the national income equation
      GDP = GOS + CE +Mixed Income

      3) btw what's Mix Income?

      4) what's Malaysia current GDP(I) breakdown?

      5) ah.. it is like what Chris rock said. (adjusted for EPL) rooney is rich but the guy who signs his cheques are wealthy

      btw, speaking about overpaid bankers in UK.. riza aziz (assuming you are familiar with his story) worked at HSBC with a CF30 Customer function (is this difficult to obtain?). what are the chances that he obtained his wealth to start a property business?

      6) this list of employee compensation is for Malaysia? I should have been an accountant.

      7) is there any breakdown for wholesalers and transport's GOS and Cost of Intermediary Goods& services?

      8) btw does this verify the claim that businesses are profiteering and subsidy is not passed to consumers in either higher salary nor cheaper price?

    19. @Zuo De

      You're welcome!


      3) Income for non-incorporated businesses e.g. sole proprietorship. This covers anything from hawker stalls to the big legal and accountancy partnerships.

      4) DOS refuses to publish them, and yes we've asked.

      5) I have no idea. I haven''t been back to the UK in 20 years. But given salary levels there, I wouldn't be surprised.

      6) Yes for Malaysia

      7) Try here

      8) I don't think profiteering is the right word, because that has the connotation of price increases and price gouging. But certainly, profits are excessive and not being shared.

      Like this for example.

    20. Thanks for the link

      4) Why DOS don't wanna share?

      7) Hmm can't find the amount on intermediary goods and services. but wow. that is a very detailed dataset

      8) there's this claim by consumers, price of sugar goes us by 20 sen, prices of drinks goes up by 20 sen (the obvious question is Milo Ais is not 100% sugar), would that constitute as profiteering? it surely feels so for me but no idea how the govt define profiteering

    21. @anon

      4) Release of the data has not been approved by cabinet (wonder why?).

      7) Try (Expenditure) - (Salaries and wages)

      8) That might be the case. However, if the price of those drinks had been held down previously, the 20sen increase might have an element of "catch up", to account for past inflation.

      There's a concept in economics called "menu costs", which means that changing retail prices is a non-trivial cost for producers, so prices don't always move in tandem with changes in input costs:

    22. 4) i cant imagine why. it is not like ppl dunno our CE is gonna low nor Mixed Income gonna be high (hawkers with BMWs and banglows)

      7) hm.. I only have sales, fixed asset, number of employees, salary paid in the dataset. no expenditure. am i missing soemthing

      8) I dunno if it can be attributed to catch up prices. last year after the petrol hike, ppl complained drink prices went up by 20 sen. when electricity tariff hike was announced, goes up another 20 sen. it feels excessive. assuming on average a restaurant sells 150 ice milo a day, that 40 sen increase is about RM1.8k extra revenue per month. if electricity bill is rm10k/month, after the 15% hike, it would increase by RM1.5k. the price hike of one item can cover the cost of the increase in electricity bill.

    23. @anon

      4) Me neither. I don't see the harm in it at all - and it might help, because then we can track what's really going on.

      7) Which particular sector are you looking at?

      8) If you look at the CPI breakdown, restaurant and hotel prices are overall keeping pace with food price increases - in some months, actually moving faster, so you may have something there. However, most of the price hikes in R&H occurred in 2011-2012, and price increases have actually slowed down in 2013.

    24. 7) wholesale and retail?

      8) will be interesting to see January 2014 numbers.

    25. @anon

      7) The 2009 Census has the data. You might also be interested in the latest numbers, though that doesn't include intermediate inputs.

      8) Yes, mighty interesting, especially the details.

  3. 1) someone playing advocate would say if we channel the resources committed to developing Felda and Petronas to non commodity industries, by the 80s we would have been more diversified to handle the crashes (if u don't look, you wont find opportunities). but admittedly, given our low skilled labour at that point (eg my parents in the 70s were the first in the family to finished schooling and went for tertiary education), opportunities to encourage manufacturing FDI were hard to find.

    5) so it is revenue neutral? that's sucky. dapat the inflation but not the extra revenue.

    6) I am in the mind that any intervention by the govt has to come from the income side (eg increase BR1M coverage and amount). let prices gravitate to market prices. and hope wages salary increase. btw does income supplementing assistance like BR1M have a dampening effect on wage growth?

    thank you for the GOS explanation. it makes sense.

    1. @anon

      1) Yes there were structural issues involved. As I said earlier, when we did get around to building the manufacturing sector, we were looking for FDI when FDI was coincidentally looking for us.

      6) No, income assistance should not cap salary/wage increases. It should also be much less wasteful than blanket subsidies.