INSIGHT
Financial Integration, Capital Controls and Monetary Independence
Renu Kohli
It has been argued for India that the increase in capital mobility has made existing controls ineffective, eroding the central bank’s influence over the short-term interest rate and that it is time, therefore, to abandon the managed exchange rate regime to regain monetary control. This note shows that the “loss of monetary independence” argument lacks empirical basis. Convergence between India and the rest of the world, though increasing, is as yet incomplete. A variety of tests shows imperfect financial integration, suggesting the Indian economy to be in the intermediate stage, i e, on the transition to full capital mobility, a feature that allows an eclectic monetaryexchange rate policy combination. With no strict trade-offs between the three policy goals posited under the trilemma, there is, as yet, no compromise on monetary independence that would justify a shift towards a free float.
Renu Kohli (renukohli@yahoo.com) was, until recently, with the International Monetary Fund.
H
But does an increase in the volume of trade – in goods, or assets or both – constitute a credible case for full convertibility? Is an increase in cross-border arbitrage opportunities conclusive proof of loss of monetary autonomy? This note shows that the “loss of monetary independence” argument lacks empirical basis. Convergence between India and the rest of the world, though increasing, is yet incomplete; in particular, the domesticforeign yield gap did not disappear even during the peak of the surge in 2007. A variety of tests shows imperfect financial integration, suggesting the Indian economy to be in the intermediate stage, i e, on the transition path to full capital mobility, a feature that allows an eclectic monetaryexchange rate policy combination. With no strict trade-offs between the three policy goals posited under the trilemma, there is, as yet, no compromise on monetary independence that would justify a shift towards a free float.
A litmus test to establish monetary independence is to verify whether capital mobility is successful in bridging the domestic-foreign market segmentation that controls are designed to achieve (Ma and McCauley 2007). To the extent that there is some kind of international arbitrage, nominal interest rates in one country will be linked to world interest rates, expectations of devaluation and risk premia. This is the concept of uncovered interest parity, which postulates that, in the absence of any taxes and restrictions, the markets will equilibrate the return on the domestic currency and the expected value of the foreign currency. So if markets are perfectly integrated, i e, no or ineffective capital controls, the domestic-foreign yield differentials should be arbitraged away by capital flows exploiting profitable opportunities. The monetary authority, in that case, would have little control over the short-term interest rate. In contrast, large and persistent offshore-onshore interest differentials would imply imperfect capital mobility. So what is the evidence on this count?
Explosion of Capital Inflows
Figure 1 (p 58) shows that the scale of foreign capital into India rose phenomenally in the last five years. As global capital flows resumed after 2001, gross capital flows rose tenfold to $ 758 bn by 2007-08; as a fraction of gross domestic product (GDP), these stood at 66%, rising fourfold from the trough of 2001-02. The net capital account more than doubled from 4% of GDP in 2004-05 to 9.5% of GDP in 2007-08. Net portfolio investments, a category of foreign capital extremely sensitive to short-term interest rates and currency arbitrage opportunities, almost trebled from $ 12 bn in 2005-06 to $ 35 bn in 2007-08. And if one were to add both trade and financial transactions with the rest of the world, India’s total external transactions as a share of GDP reached 117% in 2007-08.
This increase in financial integration was also accompanied by financial market
Economic & Political Weekly
EPW
INSIGHT
development, domestic as well as abroad. Particularly relevant to the issue of monetary control, offshore markets sprang up,
Figure 1: India – Capital Flows
800
700
600
500
400
300
200
100
01992 1994 1996 1998 2000 2002 2004
linking the domestic and the offshore markets through arbitrage and hedging activities. Thus, new arbitrage opportunities opened up besides the conventional trade misinvoicing channel, leading some to hastily conclude that capital account restrictions in India have lost their bite. A typically cited example is of the active rupee derivatives market in Singapore, which allows investors to get an exposure to Indian interest rates and the rupee-dollar exchange rate almost as they would onshore. To be sure, when a price move occurs in one market, it also gets transmitted to the other as market participants rebalance their risk exposures. The comovement in the rupee-dollar spot, forward
Gross Capital Account ($ bn) Net Flows (% GDP) Gross flows (% GD) towards more opportunities for evasion of capital controls, the question about their efficacy pivots upon the volume and persistence of capital move
-70 ments. Clearly, the scale of cross-border flows must be of
- 60
a level and endurance so as
- 50
to equalise the domestic-for
- 40
eign currency yields in order
- 30 for the monetary authority to lose control over the domes
- 20
tic interest rate. The key
- 10
questions therefore are: how
-0
big are the volumes traded, i
2006 2007
e, the relative sizes of the offshore-onshore markets? Is the arbitrage successful in eliminating the onshore-offshore yield gap? Is convergence, if it obtains, a temporary breach, or does it persist?
The size of the offshore market is not as large as the onshore markets; turnover (the amount traded) in the offshore currency market is thought to be around $750 million daily, compared to $2-3 billion daily in the onshore forward currency market. Almost 80% of the daily trading turnover in the rupee is onshore; obviously, the development of the offshore markets is limited by the development of onshore markets and the extent of controls on cross-border transactions, as these restrict the amount of hedging activity that can
be done onshore.
Figure 2: Spot, Forward and Non-Deliverable Forward Rates (1 Year, Re/$)
62
58
54
50
46
42
38
9/05 1/06 5/06 9/06 1/07 5/07 9/07 1/08 Source: Bloomberg, LP
and non-deliverable forward (NDF) rates indicates exploitation of price differentials by market traders (Figure 2). For example, a market participant exposed to a rise in the rupee against the dollar in the offshore market (a “long position”) might “go short” in the domestic derivatives markets by selling rupees forward.
Though larger volumes of external flows and greater market integration do point Trading volumes have also been driven by cyclical and temporary factors in recent years: Gyntelberg and Remolona (2007) report that leveraged investors concentrated heavily upon the Indonesian, Indian and the
5/08 9/08 1/09 3/09
Philippines currencies
during 2004-07 and the rise in the offshore market turnover partly reflects leveraged trades, with short-investment horizons and potential returns on carry trades.
Cross-Border Arbitrage?
The structure of the market too indicates the potential of cross-border arbitrage. The distribution of transactions shows that non-residents’ participation is limited to 10% of the daily forex turnover in the onshore market; these are more active in the offshore non-deliverable market that has developed in response to controls, causing trading activity to segment. That arbitrage opportunities are relatively low is further reflected by India’s place at the lowest range of nonresidents’ participation in the trading turnover of Asian currencies; the share of financial customers in its total daily forex turnover is also the lowest in Asia at 17%. Trade in goods and services is also the major driver of activity in the Indian currency markets with the
majority of transactions between exporters and importers, note Tsuyuguchi and Wooldridge (2008). The low degree of capital mobility is further observed in the low turnover of the rupee relative to underlying economic activity: the ratio of foreign exchange turnover to trade flows is about 12, less than half of the global average where turnover in foreign exchange markets is 30 times more than trade in goods and services.
How easy is it for residents, the major participants in the foreign exchange market, to arbitrage through overseas markets? Barring disguised current account transactions, e g, trade misinvoicing, the routes for domestic investors to arbitrage through overseas markets, though growing, are still quite limited.
A popular route in recent times has been “overseas loans” by resident Indian firms. In the recent capital inflow surge, foreign capital entered into India largely through the portfolio capital (non-resident, equity participation) and “overseas loans” (foreign debt, residents) routes (Figure 3, p 59). External loans by residents doubled to 3.4% of GDP between December 2006 and March 2007 quarter and remained at these elevated levels until March 2008, coming off sharply due to the sudden deterioration in the external environment. Though some of the external borrowings financed the strong growth in domestic investment, the sudden increase and ebb suggests this as a possibly leaky route for arbitrage by residents (carry trade). Domestic investors are reported to have raised foreign currency loans abroad, depositing the converted rupees into domestic accounts through illegal routes.1 The temporary rise in NDF trades during
Spot rate 12 month forward 12 month NDF June 27, 2009 vol xliv nos 26 & 27
INSIGHT
FIgure 3: Net Capital Flows – Types (% of GDP)
14
12
10
8
6
4
2
0
-2
3/2005 9/2005 3/2006 9/2006
-4 FDI Portfolio Loans
-6
volatile market conditions is also attributed to resident market participants with tradelinked foreign currency exposures.2
Effective Controls
The evidence thus far appears to suggest that the existing capital controls are having the intended effect. Trading is segmented between onshore and offshore markets: forex swaps are predominantly
Figure 4: Onshore-Offshore Yield Spreads and Policy Rates (in %)
10 9 -
Repo Reverse repo 2.22 1.65 1.34 2.24 4.62
2004 2005 2006 2007 2008 Source: Bloomberg, CEIC and authors calculations.
traded onshore while non-deliverable forwards are traded offshore. As the name suggests, all deals in the NDF market are forward deals settled in dollars. It’s not a spot market since the rupee, a non-convertible currency, cannot be ‘delivered’ on the offshore market. Opportunities for speculation are limited through restrained growth of the derivatives markets; capital controls are able to influence the level and composition of derivatives activity. Such segmentation ensures the effectiveness of foreign exchange controls, though undeniably restraining the liquidity of foreign exchange markets. Elsewhere, The Reserve Bank of India reports that the NDF volumes are not large enough to affect the domestic onshore market in normal conditions,
3/2007 9/2007 3/2008 9/2008
Banking capital Other capital
though they do impact on the domestic spot market during periods of volatility (RBI 2007).
The segmentation is evident in the persistence of onshore-offshore yield gaps between one year government paper and the NDF implied yields (Figure 4).3 On an average, the onshore-offshore yield differential at one year maturity has ranged between 222 and 462 basis points throughout the last five years, with a minimum average spread of 134 basis points in 2006. NDF prices not just embody information about market expectations of potential pressures upon the exchange rate, but also indicate positioning by investors, which is synchronised with the direction of cross-border capital movements. For instance, the widening yield gap in 2007 indicates not just domestic monetary tightening4 but also a substantial fall in offshore rates, which, in turn, reflect offshore positioning in anticipation of a strengthening rupee. And it is not a coincidence that the pace of reserve accumulation accelerated at this time, growing 32% in 2007 against 7% in 2006, as capital inflows continued to swell.
Did onshore-offshore yields converge, albeit temporarily, with the high degree of capital mobility observed in 2007, when total portfolio inflows nearly quadrupled over the previous year’s level?
Zeroing in on this particular episode, Figure 5 (p 60) uses daily data to compute the onshore-offshore spread. This shows a persistent yield gap of at least 100 bps, indicating continuous market segmentation. During this period, capital flows were routinely unidirectional – inwards, with an extraordinary surge. Indeed, an unprecedented $5.5 bn of foreign portfolio investments rushed into Indian markets in July 2007, exceeding the cumulative inflows of
4.8 bn for the entire year, January-June 2007, in one month alone! But as the chart shows, the onshore-offshore spread only briefly fell below 200 basis points in July-August 2007 touching a low of 118 bps even though the NDF-implied yields fell substantially below onshore yields during this period.
The evidence thus shows that the existing cross-border restrictions clearly prevent arbitrage of the kind that would dissolve the onshore-offshore yield gap. Even when capital inflows peaked to all-time highs in 2007, the sustained surge was insufficient to arbitrage away the onshore-offshore yield gap even temporarily.
Standard tests of uncovered interest parity provide further evidence of the efficacy of capital controls. If interest rate differentials converge to some equilibrium level determined by a country’s risk considerations, then the series should be stationary. The movement of the uncovered interest parity relationship over time, or its stationary
| Review of Labour | |
|---|---|
| May 30, 2009 | |
| Beyond the Factory: Globalisation, Informalisation of | – Kalyan Sanyal, |
| Production and the New Locations of Labour | Rajesh Bhattacharyya |
| Neoliberal Subjectivity, Enterprise Culture and New Workplaces: | |
| Organised Retail and Shopping Malls in India | – Nandini Gooptu |
| The Effects of Employment Protection Legislation | |
| on Indian Manufacturing | – Aditya Bhattacharjea |
| Power, Inequality and Corporate Social Responsibility: The Politics of | |
| Ethical Compliance in the South Indian Garment Industry | – Geert De Neve |
| Revisiting Labour and Gender Issues in Export Processing Zones: | – Mayumi Murayama, |
| Cases of South Korea, Bangladesh and India | Nobuko Yokota |
| Defragmenting ‘Global Disintegration of Value Creation’ and Labour Relations | – G Vijay |
Economic & Political Weekly
EPW
INSIGHT
properties, will thus reflect whether deviations from the mean are transitory or permanent (Cheung et al 2003). What is the evidence for India in this regard?
Figure 5: The Onshore-Offshore Yield Gap in 2007 (in %)
4.0
and help restore parity. But the existing capital
3.0
controls in India still prevent his kind of ar
2.0
bitrage. Two, the point estimate of the autore
1.0
gressive coefficient is
1/3 12/4 24/5 5/7 16/8 27/9
Unit root tests of uncovered interest differentials at overnight, one month and three months horizons strongly reject stationary in levels indicating that shocks to uncovered interest parity are permanent, or that parity does not revert to its equilibrium value after a disturbance. In other words, there is no in-built equilibrating mechanism, viz, capital mobility, to restore the parity condition and lead to convergence. Another way of answering the same question is to look at the speed at which the equilibrium interest differential, i e, the tax equivalent of capital account restrictions and the country risk premium, converges. In principle, this will depend upon the degree of development of the domestic capital market and the degree of capital mobility. Higher the restrictions, slower will be the speed of adjustment of deviation from the equilibrium; alternately, if capital is able to move relatively freely, we should observe a more rapid convergence towards equilibrium.
Assuming that the interest rate differential follows a univariate process, a significant autoregressive process would indicate capital controls, with a coefficient magnitude closer to one; the opposite would be true if the differential converges to zero with a small or zero coefficient. The results from the regressions provide interesting insights into the process of convergence.5 The coefficient is statistically significant and displays strong persistence up to five lags at all short-term horizons (except one month). This indicates imperfect integration – unsurprising because India still retains significant capital controls – indicating that deviations from the parity condition are predictable.
So markets are not efficient enough to arbitrage away the deviations as yet; if capital moves across markets quite freely, arbitrage can generate profits based upon a predictable pattern of persistent deviation
8/11 20/12
also close to unity, signifying the existence of country risk and capital restrictions. Three, the overnight market interest rate differential (interbank call rate less the US federal funds rate) has the smallest AR(1) coefficient, 0.39, over 2000-07, converging close to zero in 2005-07, perhaps reflecting increasing financial market linkages at least at very short horizons.
Straightforward regressions that test for uncovered interest parity between India and the rest of the world further reaffirm this interpretation.6 Testing for the hypothesis that the rupee-US dollar interest differentials are significantly different from zero at various horizons (overnight, one month and three months) for the period 2000-2009, we find that the average value of the differential ranges between 4 and 6 percentage points; the trend, however, is significantly declining. Has the magnitude of the deviation declined with rising financial integration? Splitting the sample into two time periods, 2000-05 and 2005-07, to account for change in monetary conditions, the mean differential is lower in the second period but remains statistically significant.7 This indicates that despite increasing financial integration, interest rate convergence does not obtain so far.
Interest Rate Divergence
Thus, despite a significant trend decline in the interest differentials in this decade, capital controls still permit the Indian interest rates to diverge significantly from those of the US Federal Reserve, notwithstanding the exchange rate linkage. Given the extremely short sample, the observed convergence prior to 2005 could also be due to convergence in inflation rates rather than weakening capital controls. The evidence only suggests that though closer financial integration is forcing convergence even when domestic monetary conditions have been tight, capital controls are sufficiently binding to provide a degree of short-term monetary autonomy to the central bank.
The facts thus do not support the hypothesis that capital controls have lost their sting with deepening financial integration in recent years. Even at the peak of the surge in 2007, when foreign capital inflows were above the trend, driven by the heady combination of a global liquidity glut and a booming economy, capital mobility was not sufficient and persistent enough to bridge the yield gap, which only dipped temporarily to nearly 100 bps, widening quickly. While some commentators concluded from this event that capital controls had collapsed in the face of the capital surge, it is clear that this was a temporary breach as some market participants exploited short-term arbitrage opportunities through existing routes. Indeed, subsequent policy responses, viz, restriction of overseas borrowings by residents through imposition of caps on amounts raised, parking funds abroad until required for expenditure and restricting automatic conversions into the domestic currency, as well as the banning of participatory notes, a derivative product, bear this out as policymakers sought to plug some channels of carry trade. These temporary opportunities arose because of market developments, rather than capital controls losing their effectiveness. In the trend case, when the net capital account ranges between 4% and 5% of GDP, we should be even less suspicious about the inefficacy of capital controls.
What inference does this evidence have for macroeconomic policy-setting in India? The evidence implies that the polar choice forced by the trilemma, to abandon exchange rate management in order to retain monetary independence, is currently not faced by the central bank to the extent made out to be. In fact, the weight of the proof demonstrates that the RBI has so far been able to choose a policy combination of part exchange rate stability and part capital account openness to operate a hybrid regime that allows it to control the domestic interest rate; consistent with
June 27, 2009 vol xliv nos 26 & 27
INSIGHT
reserve accumulation, this is critical to sustaining monetary independence. The polar tradeoffs of the trilemma do not really apply to India at this point as the economy is still in the transition stage to fuller openness, which allows an intermediate policy-mix. This enables non-linear policy adjustments as capital controls are still in place and somewhat effective, despite the fact that some cross-border flows respond to price signals in an increasingly open economy.
A recent study, Aizenmann et al (2008), is particularly illustrative in this context as it establishes that all three variables of the trilemma, i e, exchange rate stability, financial integration and monetary independence, have converged for developing countries. This, according to them, suggests that these countries “…have converged towards managed exchange rate flexibility buffered by sizeable international reserves holdings, enabling the retention of monetary autonomy even as financial integration proceeded.”
Their study also shows that the coefficients on the three macroeconomic policy goals vary over time, suggesting countries alter the weights on the three policy goals. Finally, the bulk of the empirical evidence on the efficacy of capital controls shows that the area where capital controls have been most successful is in providing more autonomy for monetary policy and altering the composition of capital inflows, while success on reducing the volume of inflows and reducing exchange rate pressures has been mixed (Clements and Herman 2009).
Notes
1 There are also other illicit routes for offshore-onshore arbitrage by residents. For instance, Indian promoters of publicly traded companies are reported to have raised personal loans offshore from large European banks against their equity holdings (shares worth at least $15 bn). The amount of such loans outstanding is reported to be $4-5 billion and the money raised was brought in through an informal remittance system that operates outside traditional banking channels (hawala) and through the accounts of non-resident Indians (see Mint, Monday, 19 January 2009).
2 The Economic Times reported on 15 September 2008, “The recent attack on the Korean won and its ripples across currency markets in Asia have turned out to be a money-making opportunity for players with overseas presence. …Multinational banks operating in India, large corporates and diamond houses have cashed in on the difference in the dollar-rupee exchange rate between the Mumbai currency market and the unregulated, unofficial, offshore markets in Singapore, Dubai and London... Corporates and institutions that have the means and flexibility have profited by buying the dollar in India and selling it on the offshore market, better known as the nondeliverable forward (NDF) market. Having undertaken such trades for years, many Indian corporates enjoy credit lines with banks abroad for NDF trades.”
3 The yield gap calculations are based upon the NDF exchange rates (1 year), offshore US dollar rates and the onshore rupee interest rates of similar maturity. based upon the methodology in Ma and McCauley (2007).
4 In January and April 2007 of 25 bps increase each. 5 Same as footnote 5, 6. 6 Regression details obtainable from the author on request.
7 The Wald test for equality of coefficients in the two time periods overwhelmingly rejects the null hypothesis.
References
Aizenman, J, M D Chinn and H Ito (2008): “Assessing the Emerging Global Financial Architecture: Measuring the Trilemma’s Configurations over Time NBER Working Paper 14533.
Cheung, Y W, M D Chinn and E Fujii (2003): “China, Hong Kong, and Taiwan: A Quantitive Assessment of Real and Financial Integration”, China Economic Review, 14, 281-303.
Clements, B, J Kamil and Herman (2009): “Are Capital Controls Effective in the 21st Century? The Recent Experience of Colombia”, IMF Working Paper No 09/30, 1 February.
Gyntelberg, J and E M Remolona (2007): “Risk in Carry Trades: A Look at Target Currencies in Asia and the Pacific”, BIS Quarterly Review, 73-82, December.
Ma, G and R N McCauley (2007): “Do China’s Capital Controls Still Bind? Implications for Monetary Autonomy and Capital Liberalisation”, BIS Working Papers, No 233.
Reserve Bank of India (2007): Report on Currency and Finance.
Tsuyuguchi, Y and P Wooldridge (2008): “The Evolution of Trading Activity in Asian Foreign Exchange Markets”, Emerging Markets Review, Vol 9, pp 231-46, December.
SAMEEKSHA TRUST BOOKS
Global Economic & Financial Crisis
Essays from Economic and Political Weekly
In this volume economists and policymakers from across the world address a number of aspects of the global economic crisis. One set of articles discusses the structural causes of the financial crisis. A second focuses on banking and offers solutions for the future. A third examines the role of the US dollar in the unfolding of the crisis. A fourth area of study is the impact on global income distribution. A fifth set of essays takes a long-term view of policy choices confronting the governments of the world.
A separate section assesses the downturn in India, the state of the domestic financial sector, the impact on the informal economy and the reforms necessary to prevent another crisis.
This is a collection of essays on a number of aspects of the global economic and financial crisis that were first published in the Economic & Political Weekly in early 2009.
Pp viii + 368 2009 Rs 350
Available from
Orient Blackswan Pvt Ltd
Mumbai Chennai New Delhi Kolkata Bangalore Bhubaneshwar Ernakulam Guwahati Jaipur Lucknow Patna Chandigarh Hyderabad Contact: info@orientblackswan.com
Economic & Political Weekly
EPW
