India's FX Reserves - Sources, Observations, Suggestions

Summary

Data indicates that RBI’s foreign currency (FCY) purchases are sourced from less permanent open exposures, rather than permanent FCY inflows. Monitoring such reversible open exposures may be an alternate way to gauge external stability, besides debt or import cover. With lower CAD and higher FDI, permanent FCY flows turned significantly positive from FY15, but could now be turning negative again. Cumulative open exposures have doubled since FY14, and also bear watching. There are possible ways to control the size of open exposures, and to make them more permanent.

A Few Observations

a) The annual and cumulative FCY purchased by the RBI across spot and forward markets is listed in column A & B of Table 1 below. These purchases can come from both permanent and transient sources. 

b) Column C & D of Table 1 provide the annual and cumulative “permanent FCY flows” – defined here as the sum of current account deficit (CAD), foreign direct investment (FDI), and foreign portfolio investment (FPI) in equity^*

c) Note that the cumulative permanent flows since FY06 is close to zero. Thus all of RBI’s $190B cumulative spot and forward FCY market purchases from FY06 to date are of FCY from less permanent sources, or “open exposures”. Open exposure (columns E & F) is a balancing figure in Table 1, and is the difference between RBI FCY purchases and permanent FCY flows. 

Period		Total RBI Purchase (Sale)	Cumulative Purchase since FY06		Permanent FCY Flows	Cumulative Permanent Flows FY06		Balance Open Exposures	Cumulative Open Exposures
		A	B		C	D		E	F
FY06		8.1	8.1		8.0	8.0		0.1	0.1
FY07		26.8	34.9		5.9	13.9		20.9	21.0
FY08		92.9	127.8		27.6	41.5		65.3	86.3
FY09		(51.6)	76.2		(25.2)	16.3		(26.4)	59.9
FY10		(0.1)	76.1		13.7	30.0		(13.8)	46.1
FY11		1.3	77.4		(6.2)	23.8		7.5	53.6
FY12		(23.3)	54.1		(38.8)	(15.0)		15.5	69.1
FY13		(10.4)	43.7		(44.7)	(59.7)		34.3	103.4
H1 FY14		(12.4)	31.3		(9.5)	(69.2)		(2.9)	100.5
FY14		(11.0)	32.7		2.7	(57.0)		(13.7)	89.7
FY15		96.2	128.9		19.9	(37.1)		76.3	166.0
FY16		(2.4)	126.5		10.4	(26.7)		(12.8)	153.2
FY17		27.5	154.0		28.9	2.2		(1.4)	151.8
H1 FY18		36.6	190.6		(5.6)	(3.4)		42.2	194.0

Table 1 – RBI FX Intervention, Permanent Flows, Open Exposures 

All figures in $Billion, source RBI Bulletin (H1 FY18 is an estimate)

* Ideally, we should consider only a part of FPI equity as permanent, rather than the whole. After all, we have seen 2 years of FPI equity outflow since FY06, and 14% of FPI equity holdings exited in FY09. However, given our presence in global equity indices, and given recent patterns, have considered the whole as permanent. 

d) While we know the possible components of open exposure, this exercise does not give us the actual split across them. Open exposures could include unhedged FPI in debt, unhedged FCY debt (including unhedged ECB & repatriable NRE deposits), net exporter forward hedges, and open trading positions in the market. Hedged FCY debt flows do not accrue to RBI’s net purchases across spot and forwards. Open trading positions would also include offshore positions that enter domestic markets, say through the merchanting trade arbitrage route. 

e) Some open exposure components are clearly more permanent than others. As shown by Acharya & Krishnamurthy (2017), restrictions on minimum tenor of ECB and minimum residual tenor of FPI debt purchases help improve the permanence of debt flows, and the quality of reserves. Nevertheless, it is worth noting that during FY09 & FY10, we did see $40B or 47% of the cumulative open exposures since FY06 flow out of the country.

  

f) Between FY11 and first half FY14, cumulative permanent outflows were almost $100B. At that point, our reserves were inadequate to cover our open exposures. 

Since then, with CAD improving and FDI rising, FY15 to FY17 permanent inflows were close to $60B, and the cumulative permanent flows moved back to zero. Cumulative open exposures therefore reside with the RBI as reserves, and this augurs well for external stability.

However, with the recent increase in our CAD, our permanent flows have turned small negative again in H1 FY18. Alongside this, there has been a sharp increase in cumulative open exposures by about $100B since FY14, much of which appear to be transient positions. Both these trends bear watching. 

A Few Suggestions

a) Traditionally, we have looked at short-term debt or import cover as a measure of reserve adequacy. However, sudden demand for FCY can come from a variety of open exposures, including unhedged imports, unhedged FCY and FPI debt (including long-term debt), trading exposures or even hedged export receivables. 

This is better captured under open exposures above, perhaps as a more holistic metric for FX financial stability than short-term debt or import cover alone. 

b) RBI has significantly increased forward FCY purchases this year, with current outstanding of over $30B. Forward FCY purchase by the RBI supplies short-term FCY debt to the market, pushes up FCY forward premia, increases system-wide open exposures, and increases the cost of reserve management.

c) If our debt was to be included in global indices just as our equity is, that could help make $70B of FPI in debt more permanent. There are possible ways of debt index inclusion that may be acceptable to both us and to FPI.

Till date, investments into equity has yielded FPI investors close to 10% IRR in USD terms, arguably a fair return for equity risk. In contrast, FPI investments in debt have provided less than that in INR terms, let alone in USD terms. With $420B of FPI equity AUM now, versus $70B of FPI debt AUM (per NSDL data) the question of proportion of portfolio debt/ equity, in relation to their relative cost, is worth considering. However, the need to ensure better permanency of FPI in debt still remains paramount, given our experience of debt outflows in June 2013.