The global financial crisis has not resulted in a general increase in EM government debt, unlike in the advanced economies. EM government debt is generally lower than in advanced economies. After all, empirical evidence suggests that emerging economies tend to suffer from so-called debt intolerance (Reinhart & Rogoff 2003). Estimates as what is a safe level of government debt typically range from 25-40% of GDP for EM, and much higher for advanced economies. Most major EM are in good shape.
Declining sovereign foreign-currency (FCY) mismatches have helped the top-tier EM avoid systemic financial crises in the past ten years. Typically, local-currency (LCY) as opposed to FCY debt affords governments with greater financing flexibility in terms of relying on the central bank as a lender-of-last resort or in terms of raising LCY liquidity through taxation. Moreover, reduced FCY debt makes government debt levels far less susceptible to sharp upward increases in the event of a balance-of-payments shock and currency depreciation.
EM total (public and private) net external debt varies greatly. Sovereign debt crises are often triggered and/ or exacerbated by the existence of contingent liabilities that governments are forced to assume (e.g. banking sector rescue). Today the EM’s total net external debt position is relatively solid. The 2008 crisis has demonstrated this already. Given restrictions on the size of FCY risks the systemically important banking sector can run, the overall FCY position is quite manageable. Combined manageable overall FCY mismatches, EM have little “fear of floating” these days in part due to the sovereign’s ability to provide FCY liquidity to the private sector and in part because “sudden stops” are self-correcting if exchange rates are allowed to depreciate. Tellingly, China and Russia, the two EM with the least flexible currency regimes, run the largest net FCY position.
External financing requirements are overall manageable and much smaller than a decade ago. More sustainable current account positions combined with larger FX reserves has translated into very manageable external financing requirements (EFR). Both the probability and the impact of a sudden stop are significantly smaller than before. Turkey, no question, is the EM most susceptible to a sudden stop. Luckily, the public sector’s net FCY exposure is close to zero, even though its FCY debt as a share of GDP remains relatively substantial at 20% of GDP. An important risk mitigant is a well-hedged banking sector. Thanks to well-hedged sovereign and banking sector balance sheets, Turkey can withstand even significant currency depreciation without running the risk of experiencing a systemic financial crisis as it used to do in the old days (e.g. 2000-01). Large EFR continue to make it susceptible to a sharp economic slowdown in the event of a sudden stop.
Non-resident holdings of EM domestic LCY government debt have increased dramatically. Low US interest rates and reduced EM sovereign FCY issuance have led foreign investors to pile into domestic LCY bonds. While this effectively transfers FCY risk from the issuer to the investor, a large share of foreign domestic bond holdings may nonetheless represent a vulnerability. Foreign holdings have proven quite stable during the 2013 tapering scare. Foreign holdings of domestic bonds are typically concentrated in longer-term, fixed-rate bonds. While this might raise volatility on this part of the interest rate curve in the event of heavy foreign selling, it also limits government refinancing risks. Short-term debt is largely held by residents and they are arguably far more likely to roll-over their debt than foreigners, even in times of market stress.
Government re-financing risks are very manageable. Given solid total external financing requirements and, even more so, the very small share of short-term FCY government debt, FCY refinancing risks have ceased to be an issue for EM. What about overall (basically: LCY) government refinancing risks? It is naturally unfair to compare the financing requirements of advanced economies to those of emerging economies. The former typically benefit from a deep and diverse investor base, often including foreign official investors in addition to a large number of non-bank investors. As far as the EM are concerned, Brazil and India, the two emerging economies with the highest government debt, have the largest gross financing requirements. India benefits from a captive investor base, while Brazil has not experienced liquidity problem with the exception of the 2002 crisis, when the prospect of Lula winning the elections and repudiating debt spooked even domestic investors. Since then, the structure of Brazilian government debt has improved dramatically. A relatively low share of foreign holdings might also be regarded as a mitigating factor in the case of Brazil. Less than 20% of domestic debt securities are in the hand of non-residents investors. In short, liquidity and foreign-currency risks attaching to government debt look quite manageable, even in Brazil and India.
 |
| Source: IMF |
Cyclically-adjusted primary balances have moved from surplus into deficit over the past few years. Commentary is often focused on this fact. The intuition seems to be that a primary deficit is equivalent to a Ponzi scheme. While in absolute terms debt does increase ad infinitum, the financially relevant metric is the debt-to-GDP ratio. As long as this ratio does not increase indefinitely, government debt is sustainable. What matters therefore is the so-called the interest-rate/ growth differential.
A favourable interest rate/ growth differential affords most EM to run a primary deficit. The IMF provides estimates of the interest-rate/ growth differentials for the EM. Brazil is the only country with a positive differential due to a combination of high domestic interest rates and low trend growth. Assuming the IMF forecasts are correct, all EM can afford to run (varying degrees!) of primary deficits without seeing their debt-to-GDP ratio indefinitely. In some cases the forecast primary deficit exceeds the interest rate/ growth differential, but the difference is small.
In the baseline scenario, the adjustment in the CAPB required to stabilise government debt falls within the margin of error (0.1-0.2% of GDP). In the case of Russia, the uncertainty attaching to the CABP forecast is especially significant given the dependence on energy-related revenues. What but if the IMF is too sanguine about the interest rate/ growth differential? This could be the case of if the IMF growth projections are too optimistic and/ or the IMF is too bearish on interest rates. The IMF growth projections appear reasonable. Only Indonesia and Mexico are projected to experience faster growth in 2014-19 than during the boom years of 2002-07. The Brazil projection may appear optimistic. But Brazil should manage to grow 2% a year over the medium-term.
In a downside scenario, government debt dynamics do not appear overwhelmingly unsustainable. If differential moves by 100 bps against the EM, most of them would need to improve their CABP. Brazil, Russia and South Africa would need to make the largest fiscal adjustment in the order of 0.5% of GDP (and sustain it) in order to stabilise debt at current levels. All said, the deterioration of the CAPB should not be too much of a concern assuming (moderate) growth projections and interest-rate projections are more or less correct. Naturally, it would be desirable for the EM to improve their CABPs in order to increase their policy flexibility and widen their fiscal space, including the capacity to react to adverse shocks.
EM government debt would be manageable, even if government were to assume significant banking sector related contingent liabilities. Two words of caution are in order, though. Calculating the CABP is as at least much an art as a science. Moreover, discretionary policy measures can quickly lead to changes in the CABP, while contingent liabilities often materialise rapidly and unexpectedly. Especially the latter can quickly add to the government burden, raise interest rates and interest payments and thereby undermine medium-term solvency, absent a broader fiscal adjustment. In addition to wars, banking sector bail-outs have historically proven a major source of contingent liabilities. Assuming, heroically, that EM governments are forced to recapitalise their banking sectors with funds equivalent to 15% of total bank lending, government debt ratios would increase, but not to an extent where it would irredeemably undermine government solvency. Luckily, EM with large banking sectors tend to have low to medium debt levels and vice versa.
All things considered, government liquidity and solvency risks, including contingent liabilities, appear very manageable in the top-tier EM. The risk arising from contingent liabilities is also manageable. Last but not least, even if EM growth were to underperform and/ or EM interest rates were to rise more than currently expected, the fiscal adjustment required to stabilise the debt-to-GDP ratio looks politically and economically feasible. High nominal/ real GDP growth and limited FCY mismatches, which allow EM to devalue their currencies in order to boost economic growth, would make such an adjustment easier to implement than in economies with a lower growth potential where currency depreciation is not a policy option (e.g. euro area).