Friday, March 29, 2013

Subsidy Rationalisation: Some Pointers From The IMF

(H/T Jason)

The cost of petrol and gas subsidies in Malaysia are likely to be in the region of RM45-50 billion this year, if not higher. Some of that comes directly from the government, while the rest are borne by Petronas or TNB.

Whether or not they appear on the government’s books, these subsidies simply have to go. First because of the actual fiscal costs of keeping the present subsidy system in place; second because of the opportunity costs for both the government and Petronas in terms of investment; and third because of the economic inefficiencies engendered by distorted prices. On top of that are the negative externalities arising from hydrocarbon use and the fact that the primary beneficiaries are corporations and the higher income households.

This is not a Malaysian problem alone, and a new IMF staff paper looks at country experiences over the past 20 years or so in dealing with fuel subsidy reductions (from the Executive Summary):


Energy subsidies have wide-ranging economic consequences. While aimed at protecting consumers, subsidies aggravate fiscal imbalances, crowd-out priority public spending, and depress private investment, including in the energy sector. Subsidies also distort resource allocation by encouraging excessive energy consumption, artificially promoting capital-intensive industries, reducing incentives for investment in renewable energy, and accelerating the depletion of natural resources. Most subsidy benefits are captured by higher-income households, reinforcing inequality. Even future generations are affected through the damaging effects of increased energy consumption on global warming. This paper provides: (i) the most comprehensive estimates of energy subsidies currently available for 176 countries; and (ii) an analysis of ―how to do‖ energy subsidy reform, drawing on insights from 22 country case studies undertaken by IMF staff and analyses carried out by other institutions.

Energy subsidies are pervasive and impose substantial fiscal and economic costs in most regions. On a ―pre-tax‖ basis, subsidies for petroleum products, electricity, natural gas, and coal reached $480 billion in 2011 (0.7 percent of global GDP or 2 percent of total government revenues). The cost of subsidies is especially acute in oil exporters, which account for about two-thirds of the total. On a ―post-tax‖ basis—which also factors in the negative externalities from energy consumption—subsidies are much higher at $1.9 trillion (2½ percent of global GDP or 8 percent of total government revenues). The advanced economies account for about 40 percent of the global posttax total, while oil exporters account for about one-third. Removing these subsidies could lead to a 13 percent decline in CO2 emissions and generate positive spillover effects by reducing global energy demand.

Country experiences suggest there are six key elements for subsidy reform. These are: (i) a comprehensive energy sector reform plan entailing clear long-term objectives, analysis of the impact of reforms, and consultation with stakeholders; (ii) an extensive communications strategy, supported by improvements in transparency, such as the dissemination of information on the magnitude of subsidies and the recording of subsidies in the budget; (iii) appropriately phased price increases, which can be sequenced differently across energy products; (iv) improving the efficiency of stateowned enterprises to reduce producer subsidies; (v) targeted measures to protect the poor; and (vi) institutional reforms that depoliticize energy pricing, such as the introduction of automatic pricing mechanisms.

You can read a Q&A by the team leaders here.

The numbers for Malaysia are sobering – pre-tax subsidies are calculated at 1.88% of 2011 GDP which is a little higher than the regional average of 1%. Two thirds of that figure is from petrol subsidies alone.

Post-tax, which includes non-fiscal costs to the economy as well as taxes not collected, the measure leaps to 7.21% or nearly 4 times higher, again with more than two-thirds coming from petrol.

The post tax figure is approximate as the IMF hasn’t yet done country specific estimates, so environmental costs for instance use a global average. Nevertheless, it’s a pretty startling difference. Just for comparisons sake, the post-tax cost is bigger than the government’s budget deficit in any year since 1987.

More importantly, the IMF staff paper includes a wide ranging, if not very deep, country case study covering 22 countries, with 14 involving petrol subsidy reform and 8 involving subsidies on natural gas. These would provide useful pointers to our own efforts, when the time comes.

Technical Notes:

  1. “Energy Subsidy Reform: Lessons and Implications", IMF Staff Paper, January 2013
  2. "Case Studies On Energy Subsidy Reform: Lessons And Implications", IMF Staff Paper, January 2013


  1. Subsidies in Malaysia is a mechanism to steal from Petronas. How else can you explain the expenditure for diesel subsidies of RM32 billion and RM24 billion for petrol?

    The 32 billion diesel subsidies is being stolen by construction and developers and by every Chong and Yeo as diesel cars are only a very small portion of vehicles. Hiding behind providing diesel subsidies to poor Malay fishermen, the Transport Ministry conveniently lumped other users.

    Can you imagine unlimited RM1 being freely given out at petrol stations nationwide? That means you and I can go and buy 40 litres diesel and sell it to a construction company, transport etc and get 40 ringgit cash! Go twice, and you get 80 ringgit cash. 10 times you get 400 cash! Every day!

    What about if my sons and daughters do it? What if my lorry tank is bigger. Yes! Free Petronas money courtesy of UMNO. But of course the brain behind is the MCA Chinese thiefs doing it for their Chinese triads out there.

    That is why the Govt close one sepet eye.

    The only karma is that diesel is carcinogen and those Chinese stealing diesel will probably die rich and of cancer!

    In the light of the nefarious reasoning for so called social subsidies all the IMF guide is really inconsequential.

    The Malay fishermen remain poor, the Chinese rich construction companies become richer unless we address and expose the obvious so that public disgust and anger will stop these rompakan which is worth billions more than those small time robbers!

  2. A bit on the hurry,joe? While any subsidy rationalization is founded on valid theoretical constructs, any hurried yanking away would yield their own social and economic costs in the short term and long term environmental gains are over exaggerated. I haven't read those arts yet so I will reserve my full comments save to say that the IMF are also motivated by other issues specially when you see the gains to be minimal post tax. And yes, what happens to the economies of oil producers? And dude don't be enraptured by the environmental hogwash for some of it ( stress some) really don't wash.

    Aargh..there it goes again, damn the net in Timbuktu.

    Warrior 231

  3. Me thought IMF is in Malaysia's bad book in term of advice?

    1. SubsidisedDieselGraduateSchemeMarch 30, 2013 at 5:18 PM

      Graduate Scheme for Diesel Tanks is now available. Instead of allowing unregulated stealing from petrol stations, the Govt has introduce subsides Diesel Tanks storage system for unemployed graduates.

      With these the graduates can go from station to station and buy subsidised diesel which they can sell to kilangs and so on.

      Billions of diesel are available and unemployed graduates should not let go of these golden opportunities of cash business. No need to go to sea or do transportation business.

      Just buy, store and sell. what could be simpler?

  4. if subsidies create distortion, how about tax incentives? corporate sector in our country is so addicted to this.

  5. It distorts resource allocation. That's it.

    1. how about tax incentives? it is a form of subsisdy and creates distortion as well. i wonder how much govt has forego its revenue.

    2. I'm not sure how much is foregone, but corporate tax + petroleum income tax is less than 10% of GDP. If we make some heroic assumptions, tax incentives might cost about half that much (or 5% of GDP).

      The difference here is that petrol use comes with negative externalities (public and social costs), while tax incentives might in fact have positive externalities. For example, double deduction on R&D spending, which I think we can all agree is probably a good thing. But some tax incentives have probably outlived their usefulness as well and need to go.

    3. Sorry, that figure should be about 2.5% of GDP, not 5%.

  6. i used to work at petrol pumps. no you cannot purchase more subsidise diesel than your vehicle's capacity can afford without being detected by the system.

    if i put in rm50 today, and tommorrow i put in another rm50, the sytem will detect and inform the stations. subsequent punishment will resort to card blocking.

    1. Free cash at diesel stationsApril 2, 2013 at 2:11 PM


      The diesel subsidy is worst because it is stealing money directly from Petronas. Subsidised diesel should not cost more than petrol. If the 32 billions were given to the poor malay fishermen, they would be millionaires by now.

      The diesel subsidy must be stopped because chinese contractors and kilangs who are not supposed to get it are stealing it.

      Direct cash to fishermen would be much cheaper but this will anger the chinese triad that MCA and UMNO are afraid of.

      Below are news clipping of fleetcard being used to smuggle 1000 litres to singapore. So much for control hehehe. You obviously dont read much do you. Organised crime syndicates may be making our so called subsidy policies. The issues on nanotag also has been going on since 2008. But someone is getting so rich, that nothing seems to be done.

      From Johor to Sabah to Sarawak every chinkee with a lorry are getting free money. Why is the Government closing their eyes purportedly because they cannot think of better ways...

  7. Part 1
    There seem to be an alarmist ring to your contention, dude...whatsup guvnor?...choking in some infernal smog in Beijing,ah?

    Now that I have read the IMF articles, the issue becomes clearer.

    Firstly, let me stress that I am all for subsidy rationalization as I mentioned in the car thread below and the economic implications of subsidies are well established. But I would caution against a) hasty b) total abolition. (a) primarily designed to prevent stressful fallout to the economy and society as a whole and (b) to be not lulled by the smug feel that we are efficient NOW to go for the total abolition of all fuel subsidy when we aint by a mile!

    As I said, withdraw the petrol subsidy (which forms the bulk of government fuel subsidy allotments) completely but no fuel tax to prevent diffuse costs being passed back to the consumer. And peg the pump prices at the prevailing ICP, to educate consumers on the rational use of fuel.

    The diesel subsidy is another issue altogether. I opine, based on data, that industry’s penchant for cheap diesel is not the sole cause for inefficiencies. The crux of the matter is a reliance on cheap labour, minimal capital investments in technology and other non-optimisations. If you remove the diesel subsidy altogether, you are going to exacerbate input costs simply because it would come as double whammy what with the recent introduction of the minimum wage thingy. Besides, there is hardly any fallback in place to tide over industry in the short term plus that costs will be dumped onto the consumer!!. Further, is the subsidy saved going to be used as a fiscal mechanism to incentivize the adoption of expensive green production modalities since the purported idea is reduction in CO2 emissions? As it is as one anon rightly don’t we have enough subsidies masquerading as incentives……. and the foregone revenues???

    I reckon it wont be used in that manner but more in direct cash handouts to the poor ala the Indonesian BLT.
    And talking about Indonesia, their modeling suggests the likelihood of “negative gains both in terms of GDP growth, diminution of per capita incomes and (surprise no surprise) detrimental impact on the lower social strata and the rural denizens as Widodo and pals indicate here (based on their use of SAM , a GEM approach): -Impact of Fuel Subsidy Removal on Government Spending.pdf

    and the Malaysian scenario is sketched along roughly similar outcomes here: 9-Economic Impacts of Subsidy Rationalization in Malaysia.pdf

    wherein the key issue being “The net result from removal depends on whether the positive impact on factor input due to technical gains and allocative effects outweigh the negative impacts of the direct removal” and how best to ameliorate the effects on the disadvantaged segments of society.

    In fact, what Najib and co should be considering in reforming the economy to be more energy efficient would be to tap into alternative sources of energy of which LNG (we have loads of it), hydroelectric power are prominent. LNG stands out as a particularly cheaper and cleaner alternative to petrol:

    Further afield, solar ( which is increasingly becoming dirt cheap), wind, offer new vistas while the thorium from the Lynas operations offer the mouthwatering prospect of clean, cost-efficient and save nuclear power ( cheap thorium reactors is the name of the game now, google to find out).

    Warrior 231

  8. Part 2

    In fact, while at the above, them planners can also shift the economic dynamics away from energy intensive manufacturing to high income growth sectors that are less energy intensive such as finance, pharmaceuticals, aerospace/robotics, services (health, education etc)/tourism etc into which the Malaysian economy is slowly but perceptibly slithering into.

    Beyond all that, despite being a global warming sceptic, I am very doubtful that certains enviro damages are reversible if they have really happened as claimed. I reckon the IMF is just creating a hooha over nothing ala the international pressies Y2K and the Mayan prophecy dupes…hahahahaaha. You know, this is IMF's backhanded strategy to goad govs to rein in unsustainable budget deficits of which subsidy slashing is a readily available bogeyman.

    Strange when they could do a a lot more by ostracizing banana republics fleecing the international financial architecture through devious means or institute standard international measures to stanch cross border bleeding via trade pricing. But then again the IMF has always been a lousy laggard in divining the impact of the rotten parts of the international financial framework on the whole, haven’t they? They even cottoned on to capital controls only after having their balls kicked bruised and blue when all they had to do was refer to the Statesman in the first place:

    Malaysian PM was ahead of his time

    Dr Mahathir Mohamad is having the last laugh. .........

    And as for the IMF figures, up to 70% or roughly 1.4t of the posttax figure would go to advanced economies leaving 600b to be divvied up by the rest of the 200 yielding some 3 billion per nation, some princely sum that is!!! Hahahahahaha...... and reducing CO2 by 13% ...what if the global poor resort to burning biomass to keep the hearth warm..some consolation ah dude? So go easy on them diesel subs, mate, for you may have the prole rabbles baying for yer blood all because of overly bothering about a few extra dimes (ROFL)

    Warrior 231

    1. Warrior,

      The IMF country case studies certainly underline the need for gradual removal, rather than shock therapy. I don't see us loping off subsidies at one go, but re-implementing what has gone before i.e. 10-20 sen removal on a quarterly or half yearly basis.

      Second, the required Pigovian tax offset doesn't add very much from my calculations, a matter of an extra 30-40 sen or so. It's not exactly calamitous, and actually a little less than the excise duty the government already foregoes. So just bringing the price back to market levels (about RM1 from the current RM1.90) plus excise duty would be sufficient. This can be done in 2-3 years at most.

      And I absolutely agree that savings from the subsidy removal should go directly to the poor. At 2012 subsidy levels, you could fund 5-6 years worth of BR1Ms from the savings of one year's subsidies.

      Apropos to your post tax comments, I noted (with some amusement) that the number one global polluter is the United States, with something like four times our emissions of greenhouse gasses on a per capita basis. The developed countries as a whole are more energy intensive than even the worst developing countries. The difference is that they mostly already tax gasoline use and we don't (although the US gas tax is below what is required).

  9. Guess that you are being tad too idealistic for your own good, man but i guess that’s the price for being Hisham ;)

    Seriously though, your suggestion of a Pigouvian tax is simply not on, simply because computing the tax rate for social costs would be near impossible and the equitability of the whole measure is very much debatable. Admittedly, this is coming from Wiki (resort to Wiki due to time constraints) but I am in sync with the views of Hayek, Buchanan and Coase as their respective perspectives makes a lot of sense and even Pigou admits as much! But I would love to delve deeper into their full take on the issue...suffice to say that with the limited ‘knowledge” at disposal I would go for regulatory methods rather than Pigovian taxation but that would bring up other issues like administrative/ enforcement costs etc.

    Secondly, I may have missed something but I think you may have glossed over an important aspect which I had raised earlier. How are the authorities going to support/incentivise industry to use more efficient and green energy modalities, since it is pretty clear that you would like any savings in subsidy to be directly transferred to the poor. or are those users of dirty fuel be left to their own devices? Given the seriousness of your initial proposal your take on this would be a much appreciated learning experience........

    Warrior 231

    1. Warrior,

      Regulation (and attendant compliance costs) are a tax in all but name. They are also less equitable, because the burden of this "tax" isn't distributed based on consumption but by merely being in the regulated industry. In that sense, a Pigovian tax is more efficient, as it more closely aligns the tax burden with actual social costs (whatever those might be). Also, regulation can also function as a barrier to market entry, which could make regulated industries more oligopolistic, but that's a separate matter.

      As far as calculating the social costs are concerned, I'll readily admit any methodology used can be challenged - but I notice that you're not arguing against the principle of it, just its implementation.

      I wouldn't normally make a push for this type of taxation, except that in the case of fossil fuel usage the argument for taxation is pretty clear. Apart from global warming (I agree with you on that - the proof that human agency is to blame is inconclusive), much of the social costs shouldn't actually be too demanding to figure out.

      As for industry incentives, I don't have any firm convictions on this issue yet. Much of the attempts made so far haven't seemed to be very effective (cap and trade for instance, or tax incentives).

      Part of the reason I think is simply inertia - Greenspan noted in his book that it took half a century for US industry to fully switch from factories designed for steam power to ones designed for electrical power (building construction and layout is completely different). The man's not great at macro, but his micro insights are pretty good.

      In this case, switching to more energy efficient methods might not be as drastic a change, but would still require a large enough investment that industries will only shift when the returns are there. In this argument, current methods are working, but we won't see the results for years. A structural adjustment incentive would then be useful in speeding up the process.

      A related problem would be that energy efficiency issues might be industry specific, in which case a global approach might not be warranted.

      Lastly, removing energy subsidies would in themselves provide an incentive for industries to become more energy efficient, whether the government provides a helping hand or not. From a policy perspective the question would then be: would industries that can't survive without an energy subsidy be worth keeping?

      As I say, I don't have any firm thoughts on this issue yet. Frankly, given your area, I'dve thought you'd have more insight into this than I would.

      And lastly, I can't believe you of all people are throwing Hayek at me...

  10. It is a million dollar question, alright and I can understand your predicament in formulating a cogent proposal simply because it is not an easy thing to solve. So if its aint broken dont fix it....hahahaha!!

    From the engineering perspective, production infrastructure is essentially linked to the form of energy source available that will result in optimized utilization of raw materials to produce the desired output at the minimal costs. Sounds economish, doesnt it but that's the reality for manufacturing is per se a profit driven activity NOT a social service. And Greenspan did not make any astounding discovery with his observation: for availability of cheap oil ensured US industry persisted with “dirty” production methods before boom...the oil shock hit and boom....the sudden rise in pro-environmental lobbyists......hahahaha

    But having said that, there is this niggling question as to why energy providers dint explore clean and relatively inexpensive alternatives like nuke reactors or natural Gas.....One reason for that is probably, costs back then was prohibitively high especially before the advent of fission reactors and cheaper “fuel” like Thorium (China has embarked on an extensive thorium-ification program...good on them). I would aver that solar energy is at its Thorium moment now: panel costs are plummeting and the use of nanotech has opened a new vista in terms of efficiency. Trust the Princeton chaps to cotton on that:

    In industries like IT for instance, internal reconfiguration of circuit architecture leads to increased energy efficiency as the IVY bridge innovation demonstrates or one could use exotic stuff like Gallium to enhance efficiencies:

    Which can also costs one, one’s life as happened in that goddamned moneylaundering hellhole, Singapork:

    But do note these gains are in final products not manufacturing process per se which brings us back to the million dollar Q above. Lets say TNB generates power via nuke reactors (which would be lower cost wise than generation via diesel powered turbines = cheaper energy for manufacturers= lower input costs= lower product price which would be a desired outcome.

    Now you contend the reason why that is not happening is because availability of subsidised diesel is a disincentive for TNB to invest in a line of thorium reactors. Solution; remove subsidies, plus nail in an enviro tax that will drive diesel costs up for TNB to consider other means of power generation and thus invest in alternatives to yield positive long term outcomes all round. But you left somethings hanging there: a) ‘in the meantime’ factor= TNB to pass on the costs to industry and industry to pass it on to consumers (you see industry does not generate the bloody thing, it consumes it and it can do only so much to optimise energy efficiency cos its configured in acertain way unless they want to yank the whole thing up and start over which will lead to costs blahblah* b) future costs = investment in thorium reactors etc etc. So you see, you have inadvertently created a problem while meaning well....hahahahaha

    *yeah...the manufacturers can recycle the energy within plant for example using heat harvesters to recover lost heat and reuse the stuff but there is again the cost thingy besides the technical adjustments

    Warrior 231

    1. Warrior,

      Absolutely right - there's a time and cost factor problem with shifting to more efficient (and less damaging) energy sources. In that case, it could be (and has been) argued that energy generation is effectively a public good and a natural monopoly. The government has to step in if the market fails to provide the desired social and economic outcome. I never saw the rationale for IPPs in the first place.

  11. Part 2

    We dueled about regulation and tax which are actually subtly different things, if my understanding of them hold. With tax: you are yielding constant revenue irrespective of use; with regulation: you only pay when you flout. So dont you think regulation is better , for a company will invest in green tech beforehand to forestall flouting and hence payment (read costs which yields revenue, or as you would put it, REALIZED TAX, for the government !!

    That is what these guys examined and they seem to imply regulations (though not too tight) triumphs:

    “The remaining question is how these findings fit into the above mentioned hypotheses regarding the impact of environmental stringency on firm behavior. As far as our country-industry-specific evidence can be inferred to firm-level behavior, the positive parameter estimates on both measures of environmental regulation lend support to the Porter and the factor endowment hypotheses. However, the negative coefficients of their quadratic terms indicate that the pollution haven hypothesis seems to hold if environmental regulation is relatively tight. After all, the evidence
    presented in this paper does not entirely confirm one of the above mentioned hypotheses on the impact of environmental regulation on investment decisions of firms.


    1. Interesting, thanks.

      But I think your understanding of regulation differs substantially from mine. I'm thinking of a tax rate, which means taxes will automatically vary with emissions, not a tax levy which won't. The removal of subsidies and/or imposition of tax on fuel would conform with this view.

      My conception of regulation on the other hand (and I could be wrong on this), is that it largely consists of compliance monitoring, and reporting costs, which are relatively fixed and won't vary with emissions, and would have to be borne by companies whether they are polluting or not.

      Certainly a fine on exceeding a given threshold would have the approximate same effects as a tax, but only above that threshold and only if the fine varied with the extent of the excess. Otherwise, I would have thought that this would act more like a fixed levy, not a variable tax, on top of the already fixed costs of compliance and monitoring.

      Of course, I have no idea what the current global practice is, or its effectiveness. More reading required here I think. Thanks for the link BTW.