Giving up the dollar addiction

There is a whiff of irony in hearing Asian government officials nagging about the potential debasement of the dollar, and then seeing them going to buy the very currency they love to hate.

Yet, that’s exactly what's been happening: Since the beginning of the year, Asian central banks have resumed their foreign exchange (FX) reserve purchases, accumulating more than $75 billion on aggregate. Peanuts by the standards of recent years, though not if one puts the number in the context of collapsing global trade and finance. Evidently, some addictions are hard to give up!

Admittedly, in some cases (e.g. Korea) the buying has been simply recuperating reserves lost in the midst of last winter’s global financial meltdown and the sudden stampede of foreign capital.

But still: At more that $5 trillion, emerging Asia’s reserves remain $870 billion higher than their end-2007 levels ($460 billion excluding India and China), making a clown out of anyone who dares suggest they are not “adequate”.

So where does one stop? Some have suggested that, as emerging markets continue to attract foreign capital flows, build up larger foreign liabilities and liberalize further their capital account, reserve adequacy considerations would justify more reserve accumulation going forward.

Frankly, I see this as a very narrow-minded argument: Reserve adequacy does not have to be achieved through reserve accumulation. But let me first discuss the framework…

The costs and the benefits: One way to assess reserve adequacy was laid out most recently in a 2008 IMF working paper titled “Are Emerging Asia’s Reserves Really Too High?” At the crux of the argument lies a cost-benefit analysis of the “optimal” level of reserves. It goes like this:

As in most things in life, when you hold FX reserves there are benefits and costs. The benefits stem from the central bank’s ability to employ those reserves in order to cushion the economy from a major disruption due to a sudden investor stampede—i.e. when foreign (and even domestic) investors decide to take their money out of the country.

The benefits of holding reserves are then larger when the probability that a country experiences a “sudden stop” of capital flows is high; and when the sudden stop can generate a very large loss of output, employment and so on, if left uncontrolled.

But reserves also involve an opportunity cost. For example, to avoid a rise in inflation, the central bank may need to absorb (or “sterilize”) the money it creates from its purchases of reserves. It will do so by selling securities domestically, on which it will have to pay interest. So when this interest is higher than that earned on the FX reserves there is a net cost.

Based on this framework, the “optimal” level of reserves is the level at which the marginal benefit from holding an extra dollar of reserves equals the cost of that extra dollar. And, per the authors’ estimations, as of late 2007, reserves in most Asian countries were more or less optimal, except for China, Taiwan and Malaysia where they were (already) unquestionably excessive.

Abstracting from any reservations one may have about some of the authors’ assumptions (and I do), there is a broader question here: How should a policymaker use this framework in order to assess (and achieve) reserve adequacy?

Reserve accumulation is not the only path to reserve adequacy: Think again of the cost-benefit framework for reserve adequacy. An obvious application of this framework is to answer the following question:

How much reserves do I need to accumulate as an “insurance” against sudden stops, given the probability of a sudden stop and given my economy’s vulnerabilities to it?

But policymakers can use the framework to tackle a different challenge: Change the “givens”!

Policy can aim, for example, at speeding up reforms that make the economy more resilient to violent moves in capital flows, in order to reduce the resulting output loss (and, hence, the need for FX reserves).

The list of reforms is long but at the top are certainly measures to increase the depth, liquidity and sophistication of local capital markets, and develop viable hedging tools (including for interest rate and foreign exchange risk) to facilitate a better management of balance-sheet risks by banks, companies and households.

Another “given” that could be changed by policy is the probability of a sudden stop. You see, while the recent stampede from emerging markets was very much driven by the shenanigans of banks in the developed world, in the past “stops” have often been driven by bad policies in the emerging economies themselves.

The opportunity cost of reserves may be higher than you think: On top of policies to change the “givens”, there is also a case for revisiting the concept of the opportunity cost of holding reserves. The IMF paper presents three different approaches to measuring this cost (of which the fiscal cost of sterilization I mentioned above is one). But there is a fourth one: The difference between the rate of return on US Treasury bills and the return from investing the money at home.

The opportunity cost seen this way will likely exceed the one measured by all three approaches in the paper. After all, high expected returns in emerging markets are the very reason foreigners enter those markets in the first place. But a higher opportunity cost means that the “optimal” reserves should be lower: A government would be better off investing part of its net export proceeds at home, instead of amassing reserves.

Finally, one would have thought that the current crisis has opened the eyes of reserve managers to another factor affecting the cost of holding reserves: The fact that reserve accumulation by many countries collectively can lead to adverse global outcomes, even when from an individual country’s perspective it might look like an optimal decision ex ante.

In other words, while it may make sense for, say, Taiwan individually to continue accumulating reserves, when everyone does it we get large global imbalances, artificially low global yields and tremendous risks to global financial stability. Risks that are not “priced in” the cost-benefit optimization framework above.

Sadly, the noises coming out of Asia betray a continued “addiction” to reserve accumulation. Even while concerns about the dollar’s future are becoming louder and louder, the solution proposed is… a new currency to park FX reserves!

In the end all the talk about the dollar’s debasement or a new global monetary system is just blabber and a distraction from the real policy challenge facing emerging markets: How to achieve reserve adequacy without relying on more reserve accumulation!