This is not an EM-wide crisis — yet
The lesson of 1997-2002 is that country-level crises may link up to become systemic
Reading the press it is clear that everyone is now an expert on emerging markets. The doom-mongers are out en masse.
I always think EM is the bi-polar asset class. We go through cycles where people either love it with a passion or hate it with a vengeance. And typically these are the same people. Maybe it’s not the asset class that is bipolar.
As so often, the reality lies somewhere between the poles.
At the start of this year, the EM cup appeared half full for most investors. Concerns over looming and further Fed tightening, the return of a strong dollar and trade wars were put to one side in the hunt for yield. As long as China and global trade, growth and commodity prices remained strong, aggregate EM could live with the higher debt service costs associated with the strong dollar and rising US rates.
This all seemed logical until the twin hits from policy errors in Argentina and continued policy failures in Turkey. The double whammy from two big, liquid markets knocked the stuffing out of EM, and the glass turned decidedly half empty. The focus came back on the Fed and the dollar. The fact that even Argentina’s IMF programme failed to change this was particularly damaging for sentiment.
What took Argentina down was positioning, and the $100bn in portfolio inflows that have flown in over the past couple of years on the consensus view that this was a turnround story — perversely, it still can be.
Turkey was perhaps different as positioning was lighter, reflecting the overheating concerns that I among many had long flagged. But the expectation was that, as on numerous occasions in the past (2006, 2013/14, 2015/16), policymakers would do the right thing, eventually creating great entry points.
But past experience failed this time around; something seems to have changed in Turkey. A Rubicon has perhaps been crossed, which may reflect a psyche change after the failed coup attempt in July 2016. As a result, Turkey’s policymakers reacted much later than in the past, with devastating consequences for the lira and Turkish markets. They can still turn this around by making the right choices but the room for error is small.
While Argentina and Turkey are big standalone stories, I am not sure we can claim that EM is heading for a systemic crisis, comparable with the period 1997-2002.
Unlike 1997-2002, big liquid EMs are characterised by floating foreign exchange regimes. Currencies are adjusting to fit, which should allow current account deficits to shrink to size for countries with large external financing requirements. Central banks are responding in an orthodox way by hiking policy rates to counter exchange rate pass through to inflation, and this should help the deflation of domestic demand to shrink external financing gaps. Inflation will be a bit higher and growth lower. There will be political consequences down the line.
While there is much focus on current account deficit EMs — including Argentina, Turkey, Pakistan and Lebanon, among others — in aggregate, EM is not running huge current account deficits. In 2017, according to the IMF’s World Economic Outlook, the aggregate CAD ex-China was just 0.7 per cent of GDP, down from 1.6 per cent of GDP in 1996 just before the onset of the Asian crisis.
Second, as EMFX adjusts, concerns understandably move to the credit space, and to countries with a weight of FX debt and with rising risks of default and debt restructuring. Yet thus far, aside from Barbados, we have not seen the kind of big EM credit event typically needed to take an EM sell-off to the level of real crisis — even Argentina and Turkey are soldiering on, albeit this may change.
This is partly explained by the relatively low ratio of gross external debt to GDP, of just over 30 per cent in aggregate across EM, according to IMF data. That said, this ratio was similar back in 1996 — and it did not help back then. But at country level a gross external debt-to-GDP ratio of 30 per cent would not raise the alarm bells. Closer to 100 per cent is where red flags are raised. EM’s are still paying.
Third, the ability to pay partly reflects the fact that despite concerns about trade wars, both global trade and real GDP growth are holding up well — and also in aggregate across EM. The IMF predicts EM real GDP growth in 2018 of about 4.9 per cent, slightly higher than in 2017.
Fourth, oil prices are holding up and still anchor at least a third of EM credits, including the Middle East, the CIS and Africa. Other commodity prices are mixed, but the bid for oil — linked to supply disruptions and to geopolitical risks, which are still elevated in the Middle East — is still holding up.
But IMF data also reveal that while debt ratios may be moderate to low, the vulnerability of EMs to a drop in global trade is rising thanks to a notable increase in their aggregate external debt service ratio, from 30 per cent in 1996 to just under 38 per cent last year. This probably reflects the fact that while gross external debt ratios are low, the nominal stock of external debt across EM is much larger than in 1996 — in fact fives times larger, at $9.5tn. Much less of this is concessionary and more is on market terms, with shorter maturities and higher service costs.
Similarly, the nominal stock of general government debt is larger, at over $15tn. As international markets have opened up, a large amount of this debt is now held by foreign institutional investors. IMF data on portfolio flows suggest an exposure across EM of around $1tn. While this aggregate number is low, at less than 5 per cent of EM GDP, such holdings tend to be heavily concentrated in favoured EM stories, including Argentina, Egypt, South Africa, Brazil, Turkey and Nigeria. By nature, these flows are “hot” and not particularly sticky. As the crisis in Argentina has proven, the dash to the exits can be quick, bringing big moves in asset prices and raising the risk of contagion.
Aggregate data do not suggest systemic top down problems looming across EM. That said, experience from 1997-2002 shows that individual country crises, often with different origins and drivers, can quickly link up to create a systemic EM theme. Back then it started with the Czech crisis, went to Thailand and Asia, then to Russia in 1998, Turkey in 2000/01, and on to Brazil and Argentina through to 2002.
The common theme then was fixed and over-valued exchange rates — not the theme now. But it is fair to say that we are seeing a weight of individual stories that at the country level are challenging and which add up to the current malaise around EM. Beyond Argentina and Turkey we have: public finances in Brazil not helped by the election cycle; sanctions risks in Russia; growth, public finances, land reform and elections in South Africa; Nafta in Mexico; long-running problems in Venezuela; and a whole host of challenges in a range of middle-level EMs — Pakistan, Lebanon, Ukraine, Ecuador, Bahrain, et al.
For sentiment to turn, one of two things must change. Either we need a critical mass of individual stories to turn, perhaps the roll out of a revamped IMF programme for Argentina, or finally the right policy response in Turkey, or just a weight of value creation from FX adjustment and higher nominal and real yields giving more carry protection.
Or, at the top-down level, we need the mood music to turn on issues such as trade and tariffs, or some shift lower in expectations of Fed tightening and the outlook for the dollar. Given Trump’s problems at home, any respite on trade seems unlikely this side of the US midterms, so the dollar looks to remain bid. At the Fed, one factor would be for concerns about EM to feed into the decision set, although this seems unlikely unless things get a lot worse in EM.
On either count, we are not there yet.