Monday, March 4, 2013

IMF Country Report On Malaysia: Reading Between The Lines

The latest Article IV Consultation between the IMF and Malaysia has some rather flattering language (excerpt):

IMF Executive Board Concludes 2012 Article IV Consultation with Malaysia

…Executive Directors commended the authorities for their skillful policies, which have underpinned Malaysia’s strong macroeconomic performance despite a weak external environment…Looking ahead, Directors considered that Malaysia’s medium term prospects are favorable, as the authorities continue to focus on safeguarding financial stability, strengthening fiscal sustainability, and securing high and inclusive growth.

Directors endorsed the current settings for monetary policy and the mildly contractionary fiscal stance in the 2013 budget. They nonetheless encouraged the authorities to further develop their medium­ term [sic] plans to restore a prudent level of federal government debt and rebuild fiscal space…

…Directors noted with satisfaction that the 2012 Financial Sector Assessment Program review found Malaysia’s financial sector to be robust, highly capitalized, and underpinned by a sound supervisory and regulatory framework…

…Directors welcomed the authorities’ intention to implement wide ranging reforms to boost growth and inclusiveness. In this context, they underscored the importance of increasing labor market flexibility, raising female participation, and fostering the skills needed for a high value-­added [sic] economy. Directors welcomed the introduction of a minimum wage, and considered that adopting unemployment insurance and pension system reforms would further strengthen social protection.

Directors took note of the staff’s assessment that, despite the significant narrowing of the current account, Malaysia’s external position appears stronger than warranted by fundamentals and desirable policies. However, they agreed with the authorities that this mainly reflects structural factors. A few Directors also pointed to underlying methodological limitations in the assessment of the external position…

I actually found the last paragraph to be  the most interesting, but that’s getting ahead of things. Overall, this assessment is fairly balanced – things are going well and policy settings are appropriate, but at the same time noting that there’s still a lot of work to do. GST, subsidy rationalisation, unemployment protection, labour market reform all get a mention. The full report, if you want to read it,  is available here.

But returning to that last paragraph though, it’s a very big shift in IMF thinking. I’ve actually been meaning to cover the underlying empirical and theoretical foundations for this, as there have been a couple of papers on this issue after the past couple of months. The IMF is also currently overhauling (warning: pdf link) its existing methodologies.

Basically what the issue amounts to is this: current methodologies for looking at over- or under-valuation of exchange rates rely on the unspoken underlying assumption that international trade is conducted with finished goods. In effect, that assumption means that exports and imports are independent of each other and supply and demand of either are regulated via differences in their domestic against external price i.e. influenced by the exchange rate.

If the exchange rate is overvalued, you would expect to see a trade deficit as imports are relatively cheaper than exports, while if an exchange rate is undervalued, you would expect to see a trade surplus as exports are cheaper than imports. If exchange rates are correctly valued, then there would be balanced trade. In practice, any current account surplus or deficit of at most 2%-3% GDP is typically taken to be “sustainable” and the exchange rate “appropriately” valued.

If however, trade is largely in intermediate goods – unfinished components or raw materials for processing into finished goods or further intermediate goods – then exports and imports are not independent and the impact of the exchange rate would be only on the domestic value-added part of the good. And we’ve seen nothing if we haven’t seen the proliferation of global supply chains and globalisation of trade links in the last two-three decades.

For example, if I export good A in USD terms, with 70% of the components sourced from imports (also priced in USD), movements of the Ringgit would raise and lower my selling price and input price in unison. The impact of the exchange rate on my sales and profits would primarily be on the 30% Ringgit-denominated and locally-sourced portion of my cost of production – in other words, the value-added.

Looking at it from the perspective of the exchange rate and the trade balance, a country with a high portion of imported content in its exports could be expected to have a high trade surplus, irrespective of the valuation or current level of its exchange rate. Conversely, countries with low import content could be expected to have balanced trade if its exchange rate is correctly valued.

In effect, what that means is that countries that have always been viewed as currency manipulators or to have had undervalued currencies, such as Malaysia, or China or Korea, actually have exchange rates that are more appropriately valued than was once thought. There is a “structural” component in their trade surpluses that existing methodologies simply don’t account for.

There’s been a burgeoning literature on global supply chains, but its been mainly in the field of management, or within economics, in terms of the impact on trade itself. The new insights on exchange rate valuation and on the balance of payments however are comparatively new.

The concept will take a long while to catch on though, as the data requirements are massive and intimidating – essentially what’s needed is a global input-output matrix, not a project to be undertaken lightly.

Fortunately the OECD and the WTO have made a start on the problem with the TIVA database which covers 40 countries. But to get a full grasp of the phenomenon we need both a broader range of countries across many more points in time.

Nevertheless the essential point needs to be made that the presence of a trade surplus or trade deficit alone is no longer sufficient in determining whether a currency is incorrectly valued. We have to look at the composition of trade and the composition of each individual good, to make an accurate and reasonable determination.

Technical Notes:

  1. Malaysia: Article IV Consultation Press Statement
  2. Malaysia: 2012 Article IV Consultation
  3. Malaysia: Financial Sector Stability Assessment
  4. Malaysia: Report on the Observance of Standards and Codes
  5. OECD – WTO Trade in Value-Added Initiative


  1. Hisham

    Is the Ringgit "incorrectly valued"?

    I don't need economists to tell me this, when my friendly neighborhood money changer tells me that my Ringgit is only worth 40-something Singapore cents!

    The indignity of it all!

    1. Jasper,

      The units of exchange are irrelevant to valuation. In the purchasing power parity terms, if the RM1.00 buys the same thing as SGD0.40, then the exchange rate is indeed correctly valued. However, valuation metrics are a bit more complicated than just purchasing power:

    2. Hisham

      With all due respect, that's the kind of gobbledygook that gives economics (and economists) a bad name!

      I'd rather go with the views of ordinary Malaysians who work in Singapore as skilled and unskilled labour. For them, every Singapore Dollar earned fetches them 2.4+ Ringgit in Malaysia.

      So, let's go easy on the "metrics"!

    3. Jasper, by your logic, the Ringgit must be stronger than the Yen or Hong Kong dollar because RM1 can buy JPY30 or HKD2.5. Do you honestly think that's true?

      All you're seeing is differences in incomes and price levels, which says nothing about whether a currency is "strong" or "weak" (which are actually used to describe over or under valuation of a currency relative to its own equilibrium level, and not its absolute purchasing power).

      Mistaking the units of currency as having some intrinsic value is known in economics as money illusion.

      Try a thought experiment - let's say the Singapore government wants to reduce the use of coinage and increase money velocity, so they introduce a new currency to completely replace the SGD and call it the New Singapore Dollar, where SGD1=NSGD10.

      Henceforth, all assets, liabilities and money flows (including wages and purchase prices) will be denominated in the new currency. If you're earning SGD10k a month now, you will be earning NSGD100k a month after the switch. A meal at a restaurant that used to cost SGD50, now costs NSGD500.

      Does the switch make you richer or poorer? Since the exchange rate now switches from SGD1=RM2.4 to NSGD1=RM0.24, does that make the RM stronger or weaker than the NSGD?

      The truth is, nothing has actually changed - money in this sense is just a unit of account.

    4. Hisham

      There's a strangely compelling logic in what you have posted, but, unfortunately, I still don't agree.

      To take your example of of the Ringgit against the Yen or the Hong Kong Dollar, I could argue that 1 Singapore Dollar fetches JPY75.3 and HKD6.21.

      It's all relative, you see. The Singapore Dollar is still intrinsically "stronger" than the Ringgit on a cross-currency exchange basis.

      But this is academic theory against real world experience.

      The real world tells me that Malaysians feel "poorer" in Singapore and that their "purchasing power" is weaker (based on anecdotal evidence from friends and business acquaintances from Malaysia who travel to Singapore regularly).

      The real world also tells me (again, based on anecdotal evidence) that Malaysians choose to work in Singapore because their earnings in SGD fetch more in MYR than if they were to work in equivalent jobs in Malaysia (unskilled labour, nurses, technicians, construction workers, security personnel....well, you get my drift).

      Let's leave this in abeyance and agree to disagree.

    5. Jasper, by all means.

      Though I have to add money in itself has never had intrinsic value, not even when it was based on metallic coinage.

    6. Hisham

      I couldn't resist asking this, since you mentioned "metallic coinage" - why are some people pushing the idea of "gold dinars"?

      And is the MYR 100 per cent backed by Malaysia's international reserves? Or is it better than 100 per cent backing?

      And are the policy makers keeping the MYR "artificially weak" so as to boost the country's export competitiveness?

      Would we better off with a "stronger" MYR and leave the exporters to reform, restructure and generally fend for themselves than relying on the crutch of a "weak" currency?

    7. Jasper,

      1. For the gold dinar, it's basically a desire for financial system stability and transactional certainty. People have forgotten that the price for that stability is real economy instability and commodity price instability.

      2. Technically, the MYR has NEVER been fully back by foreign reserves. This is partly semantic - the Ringgit was only officially introduced in 1975. The last time its predecessor, the Malayan Dollar, was fully backed by foreign reserves was in 1967 when Malaysia switched from a currency board system to a central banking system under Bank Negara.

      Very few countries currently operate currency board systems - Hong Kong is the only major rich economy that continues to do so, and Brunei is the only other country in this region.

      3. This goes back to the question of what weak or strong means, and the whole point of this blog post. Read the first sentence of the last paragraph of the IMF statement - the IMF staff assessment is that the trade balance is too much in surplus, ergo the Ringgit must be too weak.

      However, all the evidence shows that BNM has been hands-off of the Ringgit since 2005 except under conditions of high volatility i.e. the Ringgit's value is market determined. Earlier IMF reports support this conclusion.

      The second point is as I wrote in the post - part of our trade surplus is due to the value-added structure of Malaysia's exports, which means that the "real" surplus (the part that should be used to determine whether the Ringgit is too weak or too strong), is considerably smaller than the gross numbers show i.e. the Ringgit is not as weak as the IMF thinks it is.

      Also, if you want to get really technical and ivory tower about this, the external exchange rate is correlated with the internal exchange rate (the exchange rate between the tradeable goods and non-tradeable goods). Development of the domestic non-tradeable portion of the economy (i.e. primarily services) will help raise both internal and external exchange rates.

      The short version: higher GDP per capita will naturally raise the purchasing power of the domestic currency, and thus cause an appreciation of the exchange rate over the long term - as long as we don't interfere in the process.

      You can point a finger for the current short term weakness in the Ringgit straight at poor commodity prices. Same thing is happening with most other commodity producers e.g. Australia, Indonesia.

  2. Is it possible for countries to consistently run trade surpluses and deficits and still be considered in long term equilibrium?

    For your hypothesis to be true, it must be possible for a currency to be "correctly valued" yet the country still run a deficit. Yet a deficit can't run forever, because you... (and I'm not very clear on this) eventually run out of money unless you start printing? In which case you will have to devalue your currency anyway. In which case the call that your currency was overvalued was more a prediction of a future devaluation rather than a judgement of "fair value" at that point in time.

    1. aetherfox,

      The generally accepted rule of thumb is that any surplus or deficit within about 2%-3% of GDP is long term sustainable and consistent with a fairly valued exchange rate - this is actually the explicit assumption built into the IMF's old valuation models.

      Nevertheless, the value-added trade hypothesis doesn't actually change the underlying economic logic about surpluses and deficits. What it does change is which exports and imports go into determining whether you are in surplus or deficit. Insofar as deficit countries are primarily importing finished goods, standard economic analysis continues to apply.

      Second, under the current financial system, you can't actually "run out of money". What you can run out of is foreign exchange to pay for imports, and that obviously can't be "printed".

      As a practical limit, the ability of a country to run a permanent trade deficit is determined by its ability to concurrently run a permanent surplus on the capital account of its balance of payments, much like the US is doing today.

      Lastly, devaluation has a precise technical meaning in economics - a change in the nominal exchange rate within a fixed exchange rate regime. Under a floating rate regime, the term used is depreciation.

      Assuming the market is efficient and the currency is floating, an exchange rate that is expected to be overvalued in the future will be sold down now, not later. This has the advantage of making the adjustment in the exchange rate both gradual rather than sudden, and of much smaller degree at any point in time.

      Under classical assumptions, downward pressure on currencies in deficit countries should have a counterpart in appreciation pressures on surplus country currencies, and this change in relative prices should even out trade imbalances over time.

      Of course we don't live in a classical economic world. All the costs of adjustment have historically fallen on deficit countries alone, where those costs have been both large and sudden under fixed exchange rate regimes. I could talk about why this happens, but this reply is getting too long.

  3. Just asking for a clarification here - you said

    "Second, under the current financial system, you can't actually "run out of money". What you can run out of is foreign exchange to pay for imports, and that obviously can't be "printed"."

    Can you actually run out of foreign exchange to pay for imports? As far as I know most imports (and causes of the deficit) are by private companies / consumers, and they don't draw upon the central bank foreign currency reserves. All they do is just buy foreign currency on the open market using local currency. As long as consumers and companies are profit and cash positive, they can continue with this practice of buying foreign currency with local currency, then paying for imports that way. The only constraint is then, as I mentioned, running out of local currency, which would... eventually... lead to depreciation as the central bank has to either print more or... I don't know.

    1. aetherfox, it's actually quite easy to run out of forex if you're not careful. The original purpose of the IMF was to manage just that - balance of payment crises.

      While its true that private companies and firms resort to the open market for their forex needs, the "open market" is essentially the domestic banking system.

      Local banks in turn source their forex assets from foreign banks; central bank forex reserves basically provide insurance if/when local banks run out of forex.

      So the real constraints, if you want to put it that way, are the combined net forex resources of both the domestic banking system and the central bank.

      Bear in mind that in our case, the Ringgit is not convertible overseas, so the only market is onshore. Also bear in mind that banking systems can be either net long (+ve) or net short (-ve) of forex. Malaysia's banks are currently net long (more assets than liabilities); Korea and Singapore are net short (more liabilities than assets).

      If the only source of infows/outflows was trade, this system would have only occasional problems. But portfolio flows are just as large and far more volatile.

      The danger is always during a crisis, rather than the normal course of business. Loss of confidence in a currency or economy leads to large and sudden capital flight, which increases the demand for forex in the banking system while simultaneously cutting off the supply. Central bank reserves need to be sufficient to cover the shortfall. Essentially, market access for both banks and private firms is shut off.

      In theory this shouldn't happen with a floating rate regime, as the local unit should depreciate with lower demand, and thus allow the market to clear at a lower exchange rate (i.e. you don't need reserves).

      Two problems here are that the depreciation itself can be very costly in both financial and economic terms (as it raises the real value of foreign liabilities while reducing the terms of trade); and it doesn't solve the issue of sudden-stop crises where the market itself actually fails.

      Again, the local money supply is almost irrelevant here. And you should be aware that more than 95% of new money creation happens within the banking system, not at the central bank.