Wednesday, April 18, 2012

IMF Upgrades 2012 Global, Malaysia Forecasts

It was just a few short months ago – January to be precise – that the IMF cut the 2012 GDP forecast for Malaysia down to just 4.0%. The latest edition of the World Economic Outlook, released yesterday, is a mite more optimistic.

The IMF now estimates the risk of a serious global slowdown to have fallen from 10% in September to about 1% now – the main area at risk is Europe, with 55% probability of a full-year recession in 2012. There’s also mention of geo-political risk centring on Iran, with a potential impact on oil supplies and prices.

Overall though, the global economy seems to be on the mend (where have we heard that before?).

There’s some nice articles included in the text – a full fledged special feature on commodity markets (the whole of Chapter 4 is devoted to this topic as well), and the behaviour of labour shares of income during and after recessions – that are worth a closer look.

If you want skip all of that, Asia (and Malaysia’s) outlook starts from pg59 of the text. In summary, growth forecast has been upgraded to 4.4% for 2012 and 4.7% for 2013, unemployment will remain below the “full” employment level, and there will continue to be a slight deterioration in the current account surplus.

One other thing worth noting is that, despite concerns over household borrowing and loan growth in Malaysia, the expansion in credit is relatively sedate compared to our peers (chart on pg13). Nevertheless, there’s a full chapter devoted to dealing with household debt which I’ll be taking a closer look at (eventually…maybe).

Technical Notes:

  1. “World Economic Outlook April 2012: Growth Resuming, Dangers Remain”, the International Monetary Fund, April 2012
  2. World Economic Outlook Database, April 2012


  1. ETP is a very important component of Malaysia's economic plan.In fact Idris Jala says its more than a plan cos of true north and detailed numbers.

    I am not an economist but have tried very2 hard to understand the ETP from a "data" viewpoint as promoted by DSIJ.

    And I am really confused cos of so many basic discrepancies.

    1. The ETP 2009 figures for population & workforce is not same as DOS numbers.Although the difference is only about 5%, it is a concern when"future planning" did not verify actual starting point,

    2. The target of USD 15k is by itself a projection.The actual threshold,I believe is set by WB on annual basis.I think its important to set a realistic number thats based on 20-30 year trends.

    I read a WB report.."China Hi-income economy 2030".And the report states that the expected growth of the High Income Threshold is 3.4%.Thus in 2020 it should be USD 18k and not USD 15k.At an exchange rate of RM would be prudent to set a local threshold of RM55k and not RM48k.(the threshold growth trend hv been pretty subdued for last 5 years,,but taking 10 yr/20yr timespan its in the 3% range)

    I am a not an economist.In my field of engineering,we tend to validate our designs and use the most conservative realistic assumptions.What is the practise for economist?Hope you can advise.

  2. @anon,

    Hope this helps:

    1. DOS conducted the latest population census in 2010, after the original NEM projections were formulated (you'll notice the forecast population growth rate figure is waaay off). In any case, the DOS population numbers themselves are projections, based on sampled birth and death rates and immigration numbers.

    2. What's realistic? There's too many unknown variables. We have on the one hand a relatively arbitrary number set by the Wrold Bank, which is adjusted based on US inflation and the purchasing power of the "international" dollar (itself an artifical construct).

    Translating that into Ringgit terms means also accurately forecasting the RM-USD exchange rate - which is also a bit of a non-starter. My guess would be that the Ringgit should appreciate on average about 2-3% per annum against the USD, based on relative incomes, interest rate differentials and relative non-trade/trade internal balance, but a guess is all this is.

    Last, there's no actual trend in the WB high income series - there's two big structural breaks in the data (1999 and 2004). Drawing a simple straight trend line through the whole sample yields a trend estimate of 2.67%. Adjusting the trend coefficient for serial correlation yields a revised estimate of 2.08%. Taking the two sub-samples where the income level was growing (1991-1999; 2004-2012) we get 3.4% and 3.9%.

    Which estimate is more realistic depends on your view of US inflation - the inflation rate for the US during the same sample period (1989-2012) is 2.8%, which is close to the un-adjusted trend estimate above. Shorter samples yield lower estimates (e.g. 2004-2012: 2.5%). Looking forward, people have forecast anything from hyperinflation to deflation in the US, so your guess is as good as mine. But this range of numbers (2.5%-2.8%) pretty much sets the target at a little above USD15k.

    The original projections under the NEM actually sets the target to be between USD15k-17k (going from memory), so USD15k is just the bottom part of the forecast target range. My expectation is that we'll actually hit that target some time in 2018-19, assuming the variables move the way I think they will, but again, that's at best an informed guess.

    Last, sorry to burst any illusions you may have, but in economics unlike in hard sciences its extremely difficult to "validate" your models, because there's almost nothing to validate against.

  3. Thats really interesting.I agree that even on BAU we should be hitting USD 15k before 2020.But it may not be the High Income Economy threshold.

    Based on the lower limit of 2.5% its USD 16k.Thus,in my opinion the more appropriate target would be USD 16k plus a 10% buffer i,e USD 17.6k.

    Based on Pemandu's assumptions of 31.6 mil population and RM 3.2 exchange rate,the minimum number is RM 1.78 trillion.

    Another aspect of the ETP that puzzles me is the EPPs of 800 bil i.e incremental GNI from the 131 EPPs;including the Business Opportunities and multipliers.These EPPs is NOT the sum total of all economics activities under the 12 NKEAs.There is life within the NKEAs outside the EPPs.

    Thus we have as per ETP ;

    1. 2009 GNI 661 bil
    2. EPPS 800 bil
    3 GNI growth for non NKEA 274 bil
    ETP 2020 GNI 1,735 bil

    Now,whats missing from the above is non-EPP NKEAs growth.

    2009 GNI of RM 661 bil with real growth of 4% should be breaching RM 1.36 Trillion in 2020..add the 800 bil EPP GNI;thats RM 2.2 trillion.

    This maybe nitpicking but I am certainly amazed at the total lack of consistency in a "Program" that so well funded and supported by Consultants & Accounts Co.

    I think USD 15k was never the real threshold.As you mentioned the lower band is 2.5% whereas USD 15k is only 1.9%.

    Thanks for the explaination.

    I have done same exercise on the wages...and there's the same type of discrepancies,

  4. On the discrepancies with DOS data:

    The fancy powerpoints was created after latest DOS data was available.The employment numbers was already published.It should have been adopted.

  5. Yes, plenty of discrepancies and you're not the first to point them out.

    What makes all this so ironic is that among the consultants for Pemandu and the NEAC is the World Bank itself.

    Personally, I think we'll probably hit the lower end of the range (in terms of total GNI), but that's ok. The trick is that population growth is well below the number they're using and still dropping, and as I said before the Ringgit should strengthen. That means we could well be above USD15k before 2020, simply because the denominator is smaller and the exchange rate more favourable.

    On the DOS employment numbers - they're statistically estimated, not hard numbers. That's why Pemandu isn't using them.

  6. Thanks point is the real target during the internal deliberations was usd17-18k.Many references to this in various docs.Even the RM1.7 trillion GNI is representative of the real threshold.
    The usd 15k is really a bail out when they did the summaries n powerpoints.Its simply that.

    Pemandu shld do better.On one hand they are talking of true North n hard numbers.On the other hand, we have to settle for discrepancies n numbers that do not compute.

    It may not be exact science but assumptions must be clear n henceforth numbers shld jell.

  7. Don't call me sir :)

    Pemandu is aware of the problem, but I don't know if solving it will be all that useful. The numbers will need to be constantly changed and revised as the situation changes. Part of the problem is that our starting points, and each milestone, will be constantly changing as well - DOS is continually revising current and historical GDP data as the underlying data firms up, so there's no telling where we actually are at any given point in time.

    It's not just the goalpost that is moving around, the ground on which you're standing on is moving too. A pretty pickle. But as you say, some consistency would be welcome.

  8. Whats the Bumi share of the GNI now and in 2020?There are interventions to ensure that the EPPs will at least progress.

    My calculations;using the same processes and data as Pemandu's is that currently we are at 26% in 2009 and will at best be 32% in 2020.But the main driver are the wages and very constant on the OS.

    But if the base assumptions i.e Pemandu's data and projections are meaningless,I would think my calcs are doubly astray.

    Bumi GNIs needs as much intervention as the national GNIs if we are to move in tandem to the true North.And resolving this will go a long way towards reducing inequality in Msia.

    Hv Teraju developed the Bumi ETP and metrics in synch with Pemandu's.Or are they working in a vacuum of policies and aspirations?

    Planning should go beyond fortune telling via a defined set of assumptions that will need to be formally reviewed and modified as we move forward.Thus the opening statement in an annual report shld essentially be a statement to confirm that base assumptions are still valid or needs to be modified.

    And Teraju should synch their own numbers accordingly.

    Its imperfect but a least a step ahead.

  9. I don't think there is as comprehensive an approach to Bumi incomes as there is on overall incomes. While I believe Teraju is indeed working in hand with Pemandu, I don't know how well their aims are integrated.

    On the subject of inequality, policies to promote Bumi businesses and incomes might reduce inter-racial inequality of incomes, but I sincerely doubt it will have much of an impact on the overall level of inequality. That's inherent in the structure of the economy itself. Improving Bumi incomes will just change the racial identity of the income stratas, not the differences between them.

  10. "Hishamh:...but I sincerely doubt it will have much of an impact on the overall level of inequality."

    Not if you attack with both wages and EBITDA prongs!

    1. Wages way behind at 28% of GDP (according to DOSM in 2005). So wages shd be restructured by reducing corporate profits/Operating Surplus. This will broadly reduced current inequality - irrespective of race.

    2. Bumi business participations in the forms of new EPPs in substantial areas, will increase their GNI.

    Btw, Teraju is simply out of their league. They can't do much for Bumis. But its good PR to showcase them around like as if the government is doing something for the Malays. They report to a Majlis that meets twice a problem-escalating governance structure, no nothing...
    Philip See's secretary probably have more clout!