On Friday in, “Correlation May Not Equal Causation, But This Divergence Looks Like Bad News [14],” we highlighted the following two charts from Credit Suisse:
As you can see, equities and global growth have tracked global FX reserves with some degree of consistency going back at least fifteen years. Now that the so-called “great accumulation [16]” has come to a rather unceremonious end thanks to slumping commodity prices, the incipient threat of a Fed hike, and China’s yuan deval, the amount held in FX war chests has decoupled markedly from stocks and economic output. That, we contend, probably doesn’t bode well.
Falling FX reserves should, all else equal, amount to a drain on global liquidity. That is, for years commodity producers were net exporters of capital, snapping up billions upon billions of USD assets to hold for a rainy day.
Well, not to put too fine a point on it, but EM now has a deluge on its hands and a long list of country-specific, idiosyncratic political factors are making the situation immeasurably worse in certain markets by putting even more pressure on local currencies (see Brazil and Turkey for instance) and hence on reserves.
To be sure, each country has its own set of problems and each situation is unique, but generally speaking, it’s time to start asking the hard questions about reserve adequacy.
While EM sovereigns as a group may be in better shape now in terms of “original sin” (i.e borrowing heavily in foreign currencies) than they were during say, the Asian Currency Crisis, the confluence of factors outlined above means no one is truly “safe” in the current environment as moving from liquidation back to accumulation will entail a sharp reversal in commodity prices and a pickup in the pace of global growth and trade.
For those curious to know which countries are running dangerously low relative to their liabilities and other important metrics, we present the following from JP Morgan.
* * *
From JP Morgan
A common metric to assess the adequacy of foreign exchange reserves is to look at external debt. Reserves of countries with a higher proportion of external liabilities in foreign currency are perceived to be more vulnerable, particularly if these liabilities have short-term maturity given rollover risks. Our colleagues in EM research highlighted that on aggregate the short term external debt profile looks manageable relative to FX reserves. Do some countries appear riskier than other?
The IMF provides the split between domestic and foreign currency debt by maturity for most countries. Typically, the majority of external debt is denominated in foreign currency. Countries with proportionally high local currency external debt (more than 30%) include India, Czech Republic, Mexico, Poland, Thailand, and South Africa. In Figure 6, we look at reserves as a proportion of both short and long term external debt, in foreign currency where available. This proportion is high for short-term debt (i.e. more than 100%) in Argentina and Turkey. Countries that appear vulnerable on total external debt including long term debt are Argentina, Turkey, Hungary, Indonesia, Poland, Mexico, South Africa, and Brazil.
The IMF proposed a framework (IMF, Assessing Reserve Adequacy, Apr 2015) to compare FX reserve adequacy across countries. In addition to short-term external debt, to capture short term refinancing needs, they augment their metric with exports, to reflect potential losses from drop in external demand or a terms of trade shock, broad money supply, to capture the risk of deposit flight, i.e. dollarization of deposits, and portfolio and other bank-related liabilities to capture the risk of capital outflows by foreign investors. We can see this reserve adequacy ratio in Figure 7, which presents the ratio of reserves to a weighted average of short-term debt, export income, broad money supply and certain foreign liabilities. Reserves in the range of 100-150 percent of the composite metric are considered broadly adequate for precautionary purposes. On their measure, countries that appear below the recommended band are Malaysia, South Africa and Turkey.
Where is dollarization of deposits a problem? As mention above the reason the IMF included broad money supply in its FX reserve adequacy metric is to capture the risk that deposits are converted into foreign currency. This is because previous EM capital account crises had been accompanied by “dollarization of deposits”. We look at the proportion of foreign currency deposits to broad money supply or M2 to see where dollarization of deposits is most extensive (with the caveat that some of these countries’ impose restrictions on foreign currency deposits). The countries that appear to hold a high proportion of foreign currency deposits relative to M2 are Turkey, Philippines, Indonesia and Taiwan. The biggest change in this ratio since last year has been in Turkey, Hungary, Malaysia, Indonesia, and Philippines (Figure 8). It is thus these countries that have been facing most intense “dollarization of deposits” currently (with the caveat that we only have data for Brazil and South Africa up until Q2).



