Does anybody know why whenever there is a discussion on foreign exchange reserve, it is typically measured in terms of imports and external debt financing?

I took these metrics for granted in the past. They seem intuitive at first. It provides a humanizing reference to what can be meaningless large numbers. But they lose their relevance after one reminds oneself of the purpose of foreign exchange reserve.

When one measured in these terms, there is an implicit understanding that the foreign exchange reserve is used for imports and external debt financing, an understanding which is false. The foreign exchange reserve is held by the central bank and the central bank is neither responsible in any payment with respect to export and import activities nor responsible in the repayment of external debts per se.

If the foreign exchange reserve is stated within the context of the ability to sustain for example 10 months of imports, does it really do so? If the importers, which is not the central bank by the way, suddenly face a liquidity crunch and unable to pay for its current and future imports, will the large foreign exchange reserve help? No, unless the central bank suddenly becomes the economy-wide importer, which diverges so far away from the traditional central banking functions.

If the foreign exchange reserve is stated within the context of the ability to finance for example 10 months of external debt, does it really do so? If private debtors suddenly go bankrupt, will the central bank take over those debts on behalf of those private debtors? No. The central bank is not a financier of private individuals or businesses.

And if the government suddenly finds itself unable to finance its debts through traditional means, either through borrowing or taxation, will the central bank step in to pay the government’s debt (or specifically, Treasury’s debt in order to clarify the fact the when one speaks of government debt, it really refers to the government sans the central bank, and the Treasury essentially manages government-sans central bank-finance)? Not really. The government or more specifically, the Treasury does not and should not have access to the foreign exchange reserve controlled by the central bank.

The central bank can of course buy treasury to help finance government deficit. But it remains that the government borrows from the central bank. The government does not take the money and pretends that the foreign exchange reserve is some kind of treasury money saved in an account at the central bank. In fact, the central bank and the government collectively do not need the foreign exchange reserve to finance government deficit in time of crisis. The central bank can do so cheaply through debt monetizing process, i.e. through the printing presses or more accurately, money created electronically.

In short, foreign exchange reserve is not government savings.

Some governments take the foreign reserve and use it to fund a sovereign wealth fund. I suppose that is a somewhat more efficient use of the money, but that is essentially an institutional abuse.

The function of the foreign exchange reserve remains for the operation of — tada! — foreign exchange, especially for a country that runs on controlled foreign exchange regime. The reserve is huge exactly to defend the currency if need be. What has happened in Malaysia is that there is a hangover from the 1990s financial crisis. Malaysia continues to have one of the biggest foreign exchange reserve in the whole world, and that is not exactly efficient.

This is especially bad when laypersons catch up with the jargon. There have been several instances in the past when a drop in foreign exchange reserve is taken by some in the public as ominous outflow of funds from Malaysia. It makes discussion on the matter completely off track and irrelevant from the very beginning, just because everybody uses the jargon and nobody understands the function of the reserve properly.

So, somebody, please, share with me why foreign reserve is typically stated in context of imports or external debt.

I think if the metrics is used to convey what I think most individuals want to convey, there is another better concept that should be used. It is the balance of payment. The measurement in terms of imports still does not fit the bill exactly, but at least, it comes closer to being accurate than measuring the foreign exchange reserve in terms of imports will ever be. But the balance of trade is not money owned by any one person. It is money owned by the whole economy.

2 Responses to “[2499] Foreign exchange reserve in terms of imports? Non! Non! Non!”

  1. on 09 Feb 2012 at 09:44 hishamh

    The metrics are more relevant under a fixed exchange rate regime, but even under a floating rate regime, they’re still useful. I agree that a BOP-based measure would probably be more appropriate, as capital flows have gotten many times bigger.

    But getting back to why the import/external debt metric is still used:

    Let’s say Company A needs to redeem USD200 million in foreign borrowing this month. What it does is instruct its bank (Bank B) to purchase USD200 million to cover the redemption. Bank B finds that it doesn’t have enough USD on its books and has to purchase those dollars in the open market. If there aren’t enough USD sellers locally (remember the Ringgit is not convertible overseas), it turns to BNM.

    The same thing happens for imports, and even more for portfolio flows.

    Central bank international reserves, in this sense, forms a liquidity backstop for the banking system. Banks can partially fund foreign exchange transactions from their own deposits, but these liabilities still need to be “covered” by matching forex assets. If the whole banking system is net short of forex, that’s where BNM’s reserves come in. If it’s net long, then there’s no problem.

    Under a fixed exchange rate regime, central bank intervention would not be discretionary – even if the banking system itself has sufficient resources, currency exchanges may result in deviations of the exchange rate from parity, which would require the central bank to buy and sell forex (accumulate or run down reserves) to maintain that parity. This leads to the assumption that changes in our external situation are directly reflected by changes in reserves. Obviously, this isn’t necessarily true under a floating rate regime like we have now.

    You’ll notice that much of the commentary on reserves has an underlying assumption that we’re still under a fixed exchange rate regime, with hopelessly screwed up notions of causality.

    A more useful measure of reserves might be net short term liabilities e.g. Ringgit-denominated foreign holdings of domestic debt and equities securities and non-resident deposit accounts. But that’s hard to figure out, or even categorise.

  2. on 17 Feb 2012 at 17:01 Hafiz Noor Shams

    Yes, I definitely agree on the point regarding fixed exchange regime and the fact we’re not in that regime purely especially given the extremely large foreign reserve is the reason I why disagree with the measurement (really, that’s the reason I wrote this in the first place).

    On the BNM backstop point, is there ever been a case of inadequate foreign currency to meet normal market operations since the (partial?) floatation of the ringgit back in 2005?

    Even if yes, I would think the banks themselves would have enough, given that exports-imports banks exchange bills/notes/whatever with each other rather than between exporters and importers directly, unless when there is a 2008 case when the banks stopped trusting each other globally.