Embrace, don’t fear, slowing accumulation of EM FX Reserves
Much has been written about the apparent fall in emerging market (EM) central bank FX reserves since mid-2014 and the dire implications for developed equity and bond markets. I think this analysis misses three important points.
First, the focus on the USD-value of FX reserves ignores significant currency valuation effects – the US dollar’s appreciation versus other reserve currencies – and thus acutely exaggerates the actual fall in reserves.
Second, the slowdown in the pace of EM FX reserve accumulation partly reflects the gradual erosion of current account imbalances between EM and developed countries and EM central banks, particularly in Asia, allowing some currency appreciation. This is arguably a positive occurrence which developed country policy makers have called for.
Finally, the quantitative easing programs in advanced economies highlight their central banks’ willingness to supplement traditional sources of bond-buying.
Emerging markets included are Argentina, Armenia, Belarus, Brazil, Bulgaria, Chile, China, Colombia, Croatia, El Salvador, Georgia, Hungary, India, Indonesia, Jordan, Kazakhstan, Kyrgyz Republic, Lithuania, Malaysia, Mauritius, Mexico, Moldova, Morocco, Peru, Philippines, Poland, Romania, Russia, Saudi Arabia, South Africa, Thailand, Tunisia, Turkey, Ukraine, and Uruguay.
I omit a few small EM economies for which monthly data is not readily available and this is therefore a slightly narrower grouping than the IMF’s EM and developing economy grouping. This does not materially change the underlying trends or my conclusions.
Basket of currencies is 60% USD, 25% EUR, 4% GBP, 4% JPY, 2% CAD, 2% AUD and 3% other; other currencies are assumed to be unchanged vs USD
Significant FX valuation effects
The USD-value of emerging market central bank FX reserves rose to about $7.1trn at end-2013, peaked at around $7.3trn in June 2014, and gradually fell to $7.0trn at end-2014 (see Figure 1). This equates to a fall in reserves in 2014 of nearly $100bn, the first annual fall for two decade. Available data suggest that the USD-value of reserves fell further in January and February (Chinese reserve data for Q1 2015 will be released mid-April).
However, these numbers and the Financial Times’ analysis ignore the rapid depreciation of most reserves currencies, including the euro, versus the USD since mid-2014. The weakening of the euro, British pound, Japanese yen and Canadian and Australian dollars cut the USD-value of EM FX reserves by about $290bn, according to my estimates (see Figure 2). If we strip out these currency moves, reserves rose by nearly $200bn in 2014 (see Figure 3). It is no coincidence that EM FX reserves peaked in June 2014, just before reserve currencies started to weaken sharply versus the USD (see Figure 1). My calculations assume that reserve managers engaged in little active currency rebalancing during 2014, as per the IMF’s latest report on the Currency Composition of Official Currency Reserves (COFER).
Embrace causes of slowing FX reserves accumulation
Despite these obvious valuation effects, the pace of EM central bank FX reserve accumulation has slowed in recent years (see Figure 3). The USD-value of FX reserves rose $765bn in 2010, four times as much as in 2014 even after taking into account currency revaluation effects.
This slowdown in reserve accumulation is partly due to i) smaller EM current account surpluses, and at the same time ii) EM central banks, particularly in Asia, becoming less proactive in neutralising these surpluses via intervention in the FX market.
Current account rebalancing not just a China and Germany story
The IMF estimates that the EM current account surplus in 2014 was the smallest in a decade at $230bn (see Figure 4). The other side of the equation of course is that the advanced economies’ deficit has shrunk.
The most prominent example is the narrowing of China’s surplus-to-GDP ratio to around 2% last year from a peak of 10% in 2007. Conversely, Germany’s current account surplus has been stable around 7% of GDP and in USD-terms has been the world’s largest since 2011. However, even if we exclude China, EM economies’ current account surplus has shrunk significantly. Similarly, even if Germany’s large surpluses are omitted, the advanced economies’ current account deficit has narrowed sharply since 2008 (see Figure 5).
Asian central banks have intervened less and allowed some currency appreciation
Figure 6 shows that central banks of EM countries running significant current account surpluses have, since the 2008 financial crisis, typically allowed current account surpluses to translate into some trade weighted  currency appreciation. Put differently they have been less proactive in neutralising these surpluses via intervention in the FX market – buying foreign currency and selling local currency and thus accumulating FX reserves. This is particularly true in Asian economies, including China, Korea, Philippines, Thailand and Taiwan.
To some extent these central banks having become more comfortable with their “precautionary” level of reserves and more sensitive to the costs associated with maintaining large FX reserves. I would point out that advanced economy policy-makers, particularly the US Treasury, have been calling for EM central banks to adopt a more hands-off FX policy stance and allow some appreciation of their “under-valued” currencies.
Note: Currencies below the line have underperformed relative to their country’s current account balance, currencies above the line have outperformed.
So EM economies’ central banks have fewer dollars to recycle into advanced market bonds and equities as they are running smaller current account surpluses (or larger deficits) and, at least in Asia, are intervening less aggressively to allow some currency appreciation. But advanced economies need fewer capital inflows from emerging markets because they are running smaller current account deficits or have larger surpluses which they can recycle into emerging or other advanced economies (see Figure 7).
Furthermore, advanced economies have in recent years been able to reduce their dependency on emerging market portfolio inflows. Central banks, including the US Federal Reserve, the Bank of Japan, the Bank of England and very recently the European Central Bank have become the bond-buyers of last resort via ambitious quantitative easing programs.
Slower pace of reserve accumulation unlikely to reverse materially
Near-term the underlying slowdown in FX reserve accumulation may abate slightly as a number of major EM economies have registered stronger trade balances. These economies’ central banks may want to neutralise some of the impact of large surpluses on their currencies by intervening in the FX market (buying dollars).
In particular Figure 8 shows that China’s annual trade surplus has surged since early 2014 to a record high of $492bn in February 2015. The improvement is broadly evenly split between higher exports and lower imports – the latter driven in part by lower commodity prices.
Long-term, I expect an ongoing albeit gradual structural shift towards greater domestic consumption in emerging markets, particularly Asia, to spur imports and keep a cap on current account surpluses and FX reserve accumulation. This will express itself through fiscal, income and interest policies but also through a somewhat greater willingness to allow modest currency appreciation. More of this in my next research note.
 Euros accounted for about 25% of FX reserves at end-2013, implying euro reserves in EM central banks of about $1.78trn or €1.3trn. After a 12% EUR/USD depreciation in 2014, euro reserves equalled about $1.56trn at end-2014 – an annual fall of over $200bn or twice the overall fall in the USD-value of FX reserves. So had EUR/USD been unchanged in 2014, the USD-value of FX reserves would have risen about $100bn in 2014. To put things in context, a 1% appreciation in EUR/USD increases the USD-value of EM FX reserves by a whopping $16bn.
 – The IMF attributes the fact that the global current account balance does not equate to zero to transportation lags, underreporting of investment income and asymmetric valuation (see https://www.imf.org/external/pubs/ft/weo/faq.htm#q3h).
 – Trade weighted index (TWI) measures a currency’s performance versus a basket of its trading partners’ currencies.
Fig 1 – IMF COFER; www.olivierdesbarres.co.uk
Fig 2 – IMF COFER; www.olivierdesbarres.co.uk
Fig 3 – IMF COFER; www.olivierdesbarres.co.uk
Fig 4 – IMF World Economic Outlook Database; www.olivierdesbarres.co.uk
Fig 5 – IMF World Economic Outlook Database; www.olivierdesbarres.co.uk
Fig 7 – www.olivierdesbarres.co.uk
Fig 8 – OECD, statistical offices, www.olivierdesbarres.co.uk