Illicit Money Flows: Where Are They?

You might remember Global Financial Integrity’s report on illicit money flows early last year, where they reported Malaysia as being the fifth highest victim of uncategorised capital outflows in the last decade.

At the time, I partially validated their findings at least in terms of the trade mispricing channel – wildly different reported values for exports and imports between different trade partners.

So since I had a bit of time, and lots of curiosity, I decided to take a bit of a deeper look into the question of Malaysian trade mispricing, using the United Nations Commodities Trade database (Comtrade), which carries data on both imports and exports as reported by nearly every trading nation (Taiwan not included).

Did I find capital outflows? Yes, I did, and on a scale that conforms to the GFI report.

But the surprising thing was that I also found evidence of substantial capital inflows as well, even after correcting for estimated shipping costs. In fact, for the past decade, capital inflows actually outweigh capital outflows (at least, if I got the flows right).

A more in depth explanation is called for here. How do you decide if money is leaving or entering the country? Let’s say you declare US$X amount of imports, but your exporter in another country declares he sent US$(X-Y) value of goods. While imports are valued at US$X, you actually pay US$(X-Y). That US$Y represents the money leaving the country.

Another way to do this is via transfer pricing. You declare US$X in terms of export value to your foreign affiliate, but he in turn declares receiving US($X+Y). Since you’re under-declaring your sale value locally, you make less “profit” and pay less corporate tax. Alternatively, your foreign affiliate is over-declaring his cost of goods, thus making less profit, and paying less corporate tax.

Notice in the latter case that it’s not necessarily money leaving the country, rather a deliberate choice of tax jurisdiction depending on which one is lower, or where enforcement is more lax.

At least, that’s how I think it goes. If anybody can think of a flaw in my logic, please let me know.

So if I’m right, local declared imports > foreign declared exports and local declared exports < foreign declared imports means either money actually leaving the country, or a reduction in corporate tax liability locally.

But that means the opposite must also be true: local declared imports < foreign declared exports or local declared exports > foreign declared imports, must mean money flowing into the country, or an increase in corporate tax liability here as opposed to elsewhere.

So here’s the raw numbers (all in USD):

	Exports		Imports
	Malaysia	World	Malaysia	World
2000	196.5	112.8	162.6	77.3
2001	176.0	104.7	146.2	68.7
2002	188.1	110.8	157.3	73.3
2003	209.4	127.5	164.9	82.1
2004	253.3	152.0	210.3	93.7
2005	283.2	169.3	228.6	102.7
2006	321.3	192.6	262.3	120.5
2007	351.9	214.7	292.2	135.6
2008	397.4	240.0	311.3	152.2
2009	314.4	191.7	247.2	123.9
2010	397.6	247.7	329.2	160.0

Graphically (USD millions):

Malaysia’s exports and imports are both systematically higher than that declared by our trade partners.

Could this be a problem at Customs or with the way DOS enumerates the data? It would seem so at first glance, but looking at some of the individual trade partners, a problem at our end should be ruled out.

To illustrate, take these two examples:

In Japan’s case, the difference in Malaysian exports and Japanese reporting of imports from Malaysia can be largely attributed to CIF costs (which aren’t included in the export numbers). CIF by the way is carriage, insurance and freight – from my limited reading, these costs vary from between 4% to 15% depending on distance, risk, availability of shipping etc. Japan’s exports to Malaysia can also be attributed to CIF, though the CIF costs are higher (using more expensive Japanese freight? possibly).

With China however you see a huge difference. China is reporting massively more imports from Malaysia than Malaysia is reporting exports to China. On the other hand the difference in China exports/Malaysian imports can probably be put down to CIF costs…until 2007. The lines amazingly actually cross after that. Both these trends together indicate export receipts not brought home, or money actually leaving Malaysia, a net loss for us.

But you can see here it’s not a systemic problem at our own borders, and more a problem with specific trade partners.

But aside from the China example above, the aggregate numbers tell a different story – locally declared exports are greater than claimed imports overseas, while locally declared imports are also greater than claimed exports overseas. In effect the two problems indicate both inward and outward capital flows.

Taking into account CIF costs (I assume a flat 10%), here’s the net flow over the last ten years (RM millions):

The total amount from 2000-2010 is about +RM340 billion. Changing the assumption over CIF costs will change the numbers, but the overall view is still that of a positive net inflow.

I haven’t gone deeper into this to work out where the problems are with specific partners except for the major ones (surprise, surprise, we have a big capital deficit with Singapore), but that seems to be something worthwhile doing.

In the meantime…shall we start talking about illicit capital inflows?

Technical Notes:

All data from the United Nations Commodity Trade Statistics Database