A Word on Q3 COFER--It Might not be What You Think

The IMF offers the most authoritative report on central bank reserves on a quarterly basis with a quarter lag.  The report, the Currency Composition of Official Foreign Exchange Reserves (COFER), covering Q318 has been released.  It may be have been overlooked during the holidays, but if and when the pundits see it, the leading takeaway is that the dollar's share of global reserves fell below 62% for the first time five years.  The end is nigh, we will be told. 

Not to worry.  It is like comparing stock market moves based on point changes rather than percentage change.  It is the wrong metric.  Because all the assets are converted into dollars, there will be some variances that are just noise, like turning your steering wheel a little before the wheels turn.  Although the data is reported to hundredths of a percent, we accept that such precision is more illusion than real and based on estimates and constantly fluctuating instruments. 

There does not seem to be a material difference between 60% or 65%. What matters is the broad developments over the years, a trend.  Over the last 20 years, the dollar's share of global reserves has been fairly stable.  The euro's share of global reserves in Q3 18 stood at roughly 20.5%.  In 1995, the German mark (15.75%), ECU (8.53%), French franc (2.35%), and Dutch guilder (0.32%) combined for just shy of 27% of global reserves.  The euro, at 20-years old, is still not bigger than the sum of its parts in terms of a reserve currency. 

The dollar's share of global reserves is not only the wrong metric, but it is also the wrong narrative.  A shift in the dollar's reserve status is simply not the main development in this space.  The most important development has been China's agreement to report its currency composition to the IMF.  China has slowly bled them in, and this is reflected in the huge shift from unallocated to allocated reserves.  Allocated reserves were valued at $10.705 trillion at the end of Q3 18, this is up from $9.643 trillion at the end of Q3 17.  While allocated reserves rose by $1.062 trillion, overall reserves rose by only $106 bln. 

Consider that countries with reserves valued at $3.505 trillion at the end of 2015 did not want to share the currency allocation of their reserves on a confidential basis to the IMF.  It was less than $700 bln at the end of Q3 18.  Given the magnitudes involved and the size of China's official reserve holdings, the process of slowly incorporating China's allocation is nearly complete.  

Before shifting gears, we should note that overall reserves fell the second consecutive quarter. This sounds counter-intuitive.  Wasn't the dollar firm in Q2 and Q3?  What seems not to be widely recognized, reserves tends to rise when the dollar is weak, not strong.  Central banks intervene and accumulate foreign assets typically when their currencies are strong,   If, on, the other hand, their currencies are weak and intervention is conduction, the reserve asset is liquidated. 

An implication of this is that overall reserves are likely to rise when the dollar is weak and fall when the greenback is strong.   This hints at some of the dynamics at work that links a strong dollar to heightened financial tensions. 

When looking at the currency allocation itself, there are three elements of change, portfolio decisions, valuations, and the reporting of China's allocation.  The dollar holdings themselves are not subject to valuation shifts.  The amount of dollars held by central banks has never been higher.  It stood at $6.631 trillion at the end of Q3 18, a $350 increase since the end of 2017.  Overall reserves fell by $43 bln over the same period.

The euro is the second most important reserve currency. Through the first three quarters of the year, central bank holdings of euros increased by $183 bln to $2.192 trillion. Most of this increase likely reflects China's admissions.  Valuation shifts may have played a minor role.  Year-to-date, the euro is off about 4.7% against the dollar. 

 The dollar value of Chinese yuan held in reserves dipped nearly half a billion dollars in Q3 to $192.538 bln.  This most likely reflects the 3.6% decline in the yuan against the dollar.  There had been a sharp rise yuan holdings earlier in the year.   Yuan reserves rose by $69 bln in H1 18.  The one thing this cannot be impacted by is China's allocation.  However,  some suspect that this could help explain what happened to Russia's Treasury holdings, which fell from $102.2 bln at the end of 2017 to $14.9 bln at the end of H2 18.  If it were a function of a central bank's portfolio allocation decision, it would be a one-off jump.  If the rise in yuan holdings is a reflection of Chinese bonds being incorporated into benchmark indices, then we can expect to see a gradual increase in the yuan's use as a reserve currency.  This is something investors will have a better sense of in the coming quarters. 

How will Brexit impact sterling's reserve currency status? Central bankers do not seem to think it matters very much.  Sterling reserves increased from $365.8 bln when the referendum was held to $480.8 bln at the end of Q3 18.   The familiarity and security associated with the Gilt market and the quantitative characteristics of sterling suggest its sustainable role as a minor reserve currency. 

Here is a context in which percentages maybe illuminating:  In 1995, three-years after the unceremonious jettison from the ERM, sterling accounted for 2.11% of the allocated reserves, less than 1/3 of the yen's share (6.77%).  At the end of Q3 18, the yen's share was a little less than 5% and sterling's share shy of 4.50%.  

One way in which the management of the yuan discourages its use as a reserve asset is by dampening the volatility.  Officials permit the dollar-yuan exchange rate to move in a narrower band than say the euro-yuan.  The low volatility means that some asset managers and investors will increase the dollar or Treasury allocation to account for China.  This helps explain why the dollar's share of reserves ought to be larger than the US share of world GDP, as many countries are still formally or informally tied to the dollar.   Sterling, by contrast, is considerably more volatile (against both the dollar and euro).   

Central bankers seem to have decided what appears to be a fairly stable allocation of reserves in a multiple currency regime.  The inclusion of China's allocation does not appear to have been much different than what other central bankers had collectively done.

Disclaimer