For much of the last few years, India has been considered a “safe haven” amongst emerging markets, reflecting dramatically-improved macro fundamentals since the taper tantrum. Against this constructive backdrop, the stress in the government bond market creates a jarring dissonance. Before the government’s announcement to reduce market-borrowing this week, Indian bonds had sold-off 120 bps since August. In comparison, the US Treasuries sold-off 65 bps, while most other emerging markets witnessed bond rallies (eg Brazil, Indonesia, Malaysia, Thailand) or modest sell-offs (South Africa, China, Korea), averaging less than 30 bps. By any yardstick, India has been an outlier.
As a consequence, India’s yield curve is steep by historical standards. Over the last eight years, the slope of India’s yield curve — the 10 Year Bond minus the Overnight Call Rate — averaged about 50 bps. In comparison, the slope tripled to 160 bps in February and March this year. The government’s borrowing programme appropriately tried to alleviate some of the stress. Net borrowing for 2018-19 has been cut by Rs 500 billion (11 per cent), issuance has been distributed more equitably across the curve, and the government will borrow 51 per cent in the first half, versus 64 per cent last year. To be sure, this creates risks down the road, as a heaver-than-normal second half borrowing will compound the back-ended borrowing of the states, in conjunction with potential fiscal slippage risks later this year. But, for now, this has succeeded in assuaging market concerns, with yields declining by 25-30 bps. Despite that, however, the slope of the curve is still 135 bps, more than twice the historical average.
The steepening of the curve imparts collateral effects. Since August, corporate bond yields have hardened by 80 bps, tantamount to three rate hikes, even as the Reserve Bank of India (RBI) cut rates in August. Furthermore, elevated bond yields necessitate a much larger primary deficit correction if Debt/GDP is to asymptote to the 60 per cent recommended by the FRBM Review Committee.
So what’s driving the sharp rise in yields? We model yields as a function of the overnight rate, bond issuance, banking sector credit growth (stronger credit demand leaves less space for banks to invest in bonds), future interest rate expectations, global rates and the exchange rate (reflecting foreign participation). The model tracks historical movements in bond yields closely. In recent quarters, however, the explanatory power diminishes meaningfully. Actual yields are meaningfully higher than implied by the model by 30-90 bps.
So what does this imply? Think of the slope of the yield curve reflecting both (a) future interest rate expectations and (b) a risk premium (that is, the compensation investors demand to hold the asset). Changing interest rate expectations are already captured in the model. Therefore the unexplained “residual” reflects the sharp increase in risk-premia in recent quarters.
The rising risk-premia, in turn, likely reflects a growing demand-supply mismatch in the bond market. On the supply front, the Centre has brought its deficit down from 4.5 per cent to 3.5 per cent of GDP over 5 years. But this has been offset by the States whose combined deficit — without UDAY — has increased from 2.2 per cent to 3 per cent of GDP. So the consolidated deficit has barely budged from 6.7 per cent of GDP in 2013-14 to 6.5 per cent of GDP in 2018-19.
In contrast, the natural/passive demand for bonds has reduced. The policymakers have progressively — and correctly — reduced financial repression by reducing the SLR ratio (the fraction of bonds that banks are forced to hold) from 25 per cent to 19.5 per cent over the last decade, so that more credit is available to the rest of the economy. Understandably, in conjunction, the “Hold-To-Maturity” component (the protection that banks received for being forced to buy bonds) has been reduced in tandem. So, banks are subject to more interest-rate risk on their investment portfolio. From a portfolio perspective, therefore, higher risk would need, ex ante, correspondingly higher returns. Put differently, as captive demand is reduced, market-clearing yields would be expected to rise. This has been compounded by bank’s mark-to-market losses in recent quarters, and their subsequent wariness to take on more risk.
In addition, “general” foreign portfolio investment (FPI) limits are almost full, and have been increased very gradually, such that foreign participation cannot increase more substantially. The policymakers are understandably cautious about opening-up FPI limits aggressively, because the benefits of increased foreign-participation (heterogeneity, liquidity) must be traded-off with the associated volatility and pro-cyclicality.
This, then, is India’s fiscal trilemma: A reduction in financial repression (reducing captive demand), without a commensurate reduction in the consolidated fiscal deficit, along with a wariness to open foreigner limits more aggressively. The trilemma creates a fragile equilibrium in the bond market, resulting in disproportionate responses to shocks.
The policy implications flow naturally. Reversing financial repression will be retrograde; instead banks must be encouraged to hedge risk more effectively. Sharply increasing FPI limits may be deemed risky and injudicious. Therefore, the first best is to reduce the consolidated fiscal deficit with states keeping up their end of the bargain. Indeed, the early read on states is not encouraging, with the combined state deficit increasing from 2.5 per cent of GDP to 3 per cent over the last 2 years.
The importance of rationalising supply is evident this week. Cutting borrowing by 0.3 per cent of GDP has already pushed down yields by 25-30 bps. The government's borrowing program did the best it could to alleviate stress. But it should be thought of as a technical, near-term solution. India’s fiscal trilemma is more fundamental in nature. And will require a commensurately fundamental response.
The author is Chief India Economist at J P Morgan Chase Bank