![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
Abstract
In view of multiple instruments used by many central banks in emerging market economies (EMEs), we derive a composite measure of monetary policy for India and assess its impact on the yield curve. Our results show that while monetary policy has the dominant impact among macroeconomic variables on the entire term structure, it is particularly strong at the shorter end and on credit spreads. Shifts in the level of the government yield curve and credit spreads also lead to changes in monetary policy. In terms of robustness, our measure performs better than a narrative-based measure of monetary policy available in the literature.
Disclosure statement
No potential conflict of interest was reported by the authors. The views expressed in the article are those of the authors and not of the institutions to which they belong. The usual disclaimer applies.
Notes
1. Burdekin and Siklos (Citation2008), Bhattacharyya and Sensarma (Citation2008) and Montoro and Moreno (Citation2011) discusses the use of reserve requirements along with policy rates in China, India and Latin America.
2. The average gross fiscal deficit of the Government, as a proportion of GDP, was 7.8% during the period 2000–2010.
3. Kanjilal (Citation2013) also calculated empirical proxies using the 3-month, 1-year and 10-year yields which, however, are not equally spaced maturities and therefore the usual curvature formula is not appropriate (see Giese Citation2008 for more on this). By using equidistant maturities, we are able to avoid this problem.
4. For instance, an increase in the repo rate accompanied by a reduction in CRR may be interpreted as confusing by the market. However, the net impact on the cost of funds would depend on the liquidity effect of higher cost of borrowing from the RBI through the repo window and greater resources available to banks from the release of impounded balances through reduction in CRR.
5. All these variables are taken in logarithmic form and IIP is also de-seasonalized by X-12-ARIMA.
6. One common approach is to take p as 4 with quarterly data and 12 with monthly data. In our case, that would leave us with limited degrees of freedom. Therefore we choose p by relying on the Schwartz information criteria which gave an optimal lag length of 1 for all the estimated models.
7. Data on both IIP and WPI come with a lag and hence cannot be used by market players in their bond trading strategies. However PCI and NEER are real-time information which have immediate bearing on bond yields.
8. We consider all variables in levels as differencing them would lead to loss of information on their relationship (Sims Citation1980; Doan Citation2000) and VAR estimates are consistent even when unit roots are present (Sims, Stock, and Watson Citation1990). In any case, our main variable of interest, that is PCI is stationary with an ADF test statistic of −2.68 which is greater in absolute terms than the 5% critical value of −1.94. Therefore, we feel it is not meaningful to take its first difference or to consider its long run relationship with the other variables in a cointegrating framework.
9. The Indian fiscal and financial year is April to March.
10. These are weighted averages of yields on all bonds of the corresponding maturity being traded in the corresponding month with the volumes as weights and yields computed on a zero coupon basis.
11. Since May 2011, RBI shifted to a new operating procedure of monetary policy whereby the repo rate became the sole independently varying policy rate while the newly constituted marginal standing facility and the reverse repo rate were pegged at 100 basis points above and below the repo rate, respectively. As our principal component measure is based on discrete and independent changes in repo, reverse repo and CRR, we have restricted our analysis to 2010 in order to avoid inconsistency in PCI measure across periods.
12. Recent empirical evidence on the operating procedure of monetary policy in India suggests that monetary transmission is stronger in deficit liquidity conditions; hence the repo rate is the key policy rate (RBI Citation2011).
13. Drakos (Citation2001) found that while monetary policy actions do exert an impact on the entire term structure in the Greek money market, the magnitude of the effect decreases monotonically with higher maturity.
14. In the finance literature, the pure expectations hypothesis assumes that all changes in yield curve steepness reflect the market’s shifting rate expectations, while the risk premium hypothesis assumes that the changes in steepness only reflect changing bond risk premia. In reality, both rate expectations and required risk premia influence the curve’s slope (Wets and Bianchi Citation2006). Taboga (Citation2009), however, cautions that more than the risk premia, it is the reduction in key variables such as the real natural rate of interest, inflation expectations and growth rate of potential output that may have contributed to a fall in long-term bond yields both in the United States and in the euro area in recent years.
15. The market’s curve reshaping expectations, volatility expectations and expected return structure determine the curvature of the yield curve. Expectations for yield curve steepening or for low volatility along with low required returns on intermediate bonds can make the yield curve convex (Wets and Bianchi Citation2006).
16. At the same time, yield spread is found to be a leading indicator of economic activity explaining about 40% of the variation in future growth of industrial output in India (Kanagasabapathy and Goyal Citation2002).
17. A positive monetary policy shock is contractionary in our study in contrast to BMI, where such shocks are interpreted as expansionary policy.
18. A value of 3 is indicative of strong contractionary, 2 is contractionary and 1 is mild contractionary policy while a value of −1 represents mild expansionary, −2 reflects expansionary and −3 represents strong expansionary policy.
19. The results are not reported here to save space but are available with the authors.
20. This is particularly true for banks as they try to hold more of government bonds not only from a risk-return perspective but also because greater holding of risk-free assets entails lower provisioning for capital adequacy.
Additional information
Notes on contributors
Rudra Sensarma
Rudra Sensarma is associate professor at the Indian Institute of Management, Kozhikode, India.
Indranil Bhattacharyya
Indranil Bhattacharyya is director in the Department of Economic and Policy Research, Reserve Bank of India, Mumbai.