The problem is that the public data aren't representing what they in reality should. The Current Account is meant to measure the net flow of money into the economy originating from net balances of goods, services and factor income (and current transfers, i.e. gifts, donations etc.). The problem is that the CA figures for Iceland are not measuring the flow, they are, to a very large extent, merely guesstimating it! And the reason: the old bankrupt banks.
Let's first look at the official figures. The green line below is the four quarter running average of the OECD figures for Iceland. The blue line is the official non-amended figures from the Central Bank of Iceland (four quarter running average as well). Obviously, those are pretty much the same figures, leading us to trust the CBI figures pre 1997 (OECD figures only extend to 1997).
Current Account balance as a % of GDP. I don't think any Western economy has ever managed to run its current account deficit as far down as Iceland did in 2007 and 2008. Figures from OECD and Central Bank of Iceland.
Iceland's current account (% of GDP) since 1946 to 2007. By 2004 we were already close to setting historical record CA deficit that should have worried most sober economists (we managed to fool us into thinking those were financed with capital inflows from private parties). The party continued for four more years, then reality knocked. Source: Statistics Iceland and Central Bank of Iceland.
Now here comes the trick: since 2008, we cannot trust the publicly announced CA figures. Since 2008, the public CA figures are not showing the actual flow that the CA figures should show.
The reason is the fact that the old banks are in receivership. The receivers of the banks are liquidating their assets and paying them out to the claimants of the old banks, normally the owners of the bonds etc. issued by the old banks. This process takes however many years and the flow of payments to the claimants, which no one knows who is (they are effectively the owners of the new banks because the new banks are, mainly, in the ownership of the receiver of the old bankrupt banks), is not constant nor can it be accurately estimated many years before it actually happens.
But due to international rules about national accounts, the Central Bank, as the accountant of the current account, must estimate what the claimants will in the end get paid. This puts the Central Bank up against the wall: how is it meant to know how much and when the claimants of the banks, which are foreign, will get?
Imagine e.g. that the assets in the receivership of Kaupthing amounts to, say, 1000 billion ISK. Those assets still earn interest and their price fluctuates while they are being liquidated. Lets imagine that the calculated interest - not flow but an estimate of the flow! - of those 1000 billion ISK are 50 billion in one quarter. This amount must be booked as negative factor income in the current account, even though no flow actually happened. The liquidation process can take years and in the meanwhile, the Central Bank must estimate the interest earned on the asset as that is accounted for as (negative) net factor income, not having the slightest clue how much worth the asset will finally be. But those estimates end up dragging the Current Account down, as it is presented, although there has been no actual flow behind the estimated interest income on the assets in the receivership. The Current Account figures of Iceland, since 2008, do not represent actual flow of money as they are meant to!
Furthermore, nobody knows how much will actually be paid out. The assets of the receiverships are fluctuating in value as they are being liquidated and once they are sold and finally paid out to the claimants, the Central Bank may have over- or underestimated the actual net factor income. Finally, we don't know when the liquidated assets will be paid out to the claimants - i.e. when they finally become an actual flow! - which is when they will, at least to some extent, register on the FX market. In short, this is a total mess!
To estimate the flow itself that is meant to be represented by the CA figures, the Central Bank has issued another quasi-public CA estimate. This estimate cuts out entirely the effects of the receiverships of the banks, and in comparison to the public figures - those issued to the OECD - they look quite different!
Once the old bankrupt banks have been cut out of the Current Account figures, they become considerably better looking.
But of course, this methodology of the Central Bank - to cut out entirely the effects of the old bankrupt banks - isn't strictly correct either. Cutting out all the effects of the receiverships of the banks ignores the flow that actually takes place due to the liquidation of banks' assets and payments to claimants.
In effect, the "true" current account deficit of Iceland is somewhere in between the two extremes. It is somewhere between -8% and -1.2%. Where exactly? No chance to know!
Hopefully, this explains some of the reasons why Iceland, according to public data from e.g. OECD and other international data banks, is still running a CA deficit amounting to such ridiculous levels that it challenges common economic sense. The situation is bad but it's not as scandalously bad as the OECD figures imply. But the CA figures of Iceland are in fact very interesting for another fact: they show extremely well the effects of bank loans on the Current Account, something I'll get into detail later.