I. Introduction

In this paper, we examine how the United States (US) bond yield affects affects foreign institutional investor (FII) inflows to India and the performance of the Indian stock market. The impact of monetary policy in the US on international financial conditions, the resultant changes in capital flows, and the differences in spillovers between advanced and emerging economies have long been studied and debated in the literature (Fratzscher et al., 2018; Joyce et al., 2011; Krishnamurthy & Vissing-Jorgensen, 2011; Zhang et al., 2020). Our hypothesis is that the US 10-year government bond and variations in its yield influence the FII inflows into emerging markets in general and India and thereby influence changes in stock market performance. This hypothesis test is important because the performance of stock markets in the emerging economies depends largely on the investments by foreign institutions.

The literature has examined the relationship between quantitative easing (QE) and stock market performance. Anand and Chakraborty (2020) study the local and international consequences of a negative interest rate policy and how it varies from QE. They find that, while QE has no discernible effect on inflation, it has a negative effect on nominal gross domestic product growth in emerging Asian markets. Bhattarai et al. (2021) examine the international spillover effects of US QE on emerging market economies. They find that the US QE shock has a considerable impact on emerging market economies’ financial variables.

Ziaei and Szulczyk (2021) examine the impact of the US monetary policy on assets, bonds, and exchange rates in a sample of East Asian nations. They find that the US monetary policy plays a substantial role in the East Asian financial markets.

A handful of studies (e.g., Bhatia & Kishor, 2013; Chandra, 2012; Choudhary et al., 2022; Marfatia, 2022; Pandey, 2020; Singhania & Saini, 2016) explore FII inflows and their impact on the Indian stock market. To our knowledge, there is no study investigating the causal relationship among US bond yield, FII inflows, and stock market performance in the Indian context. Our study attempts to fill this gap in the literature.

The remaining paper is organised as follows. Section II discusses the methodology. Section III discusses the results, while Section IV concludes the paper.

II. Research Design and Methodology

We use monthly time-series data ranging from 2002 to 2021. To test the interrelationship among US bond yield, FII inflows, and stock market performance, we use a vector autoregressive (VAR) model and the Granger causality test.

A. Data

We collect monthly historical data on the 10-year US bond yield from www.investing.com. The monthly FII inflows (net investment) data are collected from www.fpi.nsdl.co.in. The monthly historical data on the Nifty index are collected from www1.nseindia.com.

B. Econometric Model

We employ a VAR model and the Granger causality test to assess the impact of bond yield on FII inflows and stock market performance. The reason for this choice is that a VAR model is flexible and is of widespread use in time-series analysis. It is mainly used for forecasting financial and economic trends, and allows users to develop real-time equation modelling. The stationarity of the series is checked by applying the Augmented Dickey-Fuller unit root test.

III. Results

A. Augmented Dickey-Fuller Unit Root Test

Panel A of Table 1 reports the Augmented Dickey-Fuller unit root test results. It is clear that the series follow a stationary process at first difference. That is, we cannot accept the null hypothesis that series contain a unit root, Thus, the time series are considered to be integrated at the first difference. Following this, we conduct the cointegration and Granger causality tests between the variables, namely bond yield and FII inflows.

B. Cointegration Test

The results obtained from the Johansen cointegration test are reported in Panel B of Table 1. The trace statistics are greater than the 5% critical value, indicating no long-run cointegration between the variables. Since the variables are not cointegrated, we employ an unrestricted VAR model to estimate their relationship.

Table 1.Unit Root Test and Cointegration Test Results
Panel A: Unit root test
ADF Test Statistics Critical Value
1%
Critical Value
5%
Critical Value
10%
Prob. Result
D BONDY -14.317 -3.996 -3.432 -3.132 0.0000 Stationary
D FII Equity -9.209 -3.467 -2.881 -2.571 0.0000 Stationary
D_NSE -15.320 -3.466 -2.881 -2.571 0.0000 Stationary
Panel B: Cointegration test results
maximum rank parms LL eigenvalue trace statistic 5% critical value
0 12 -4019.1084 . 312.5347 29.68
1 17 -3952.868 0.43506 180.054 15.41
2 20 -3898.4677 0.37435 71.2534 3.76
3 21 -3862.841 0.26444
maximum rank parms LL eigenvalue max statistic 5% critical value
0 12 4019.1084 . 132.4807 20.97
1 17 -3952.868 0.43506 108.8006 14.07
2 20 3898.4677 0.37435 71.2534 3.76
3 21 -3862.841 0.26444

This table presents Augmented Dickey-Fuller (ADF) unit root test and Johansen cointegration test results in Panels A and B, respectively. NSE represents the Indian stock market; FII_EQ refers to the FII inflows to equity market, and the BONDY refers to the US Bond Yield. D represents the first-order difference to the time series.

C. The VAR Model

Table 2 shows the appropriate lag lengths based on the Likelihood ratio (LR), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC), and Hannan–Quinn Information Criterion (HQ). The SIC suggests no lag, whereas the HQIC, AIC, and LR suggest, respectively, lags of two, six, and seven.

Table 2.Lag Order Selection Statistics
lag LL LR df p AIC HQ SIC
0 -3829.64 33.9171 33.9355 33.9625*
1 -3819.47 20.335 9 0.016 33.9068 33.9801 34.0884
2 -3782.77 73.387 9 0.000 33.6617 33.79* 33.9796
3 -3776.69 12.173 9 0.204 33.6875 33.8707 34.1416
4 -3773.22 6.9282 9 0.645 33.7365 33.9747 34.3268
5 -3756.4 33.648 9 0.000 33.6673 33.9604 34.3937
6 -3738.83 35.148 9 0.000 33.5914* 33.9395 34.4541
7 -3730.04 17.578* 9 0.040 33.5932 33.9964 34.5922
8 -3726.92 6.2271 9 0.717 33.6453 34.1034 34.7805

This table presents the lag order selection statistics based on the Likelihood (LR), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC), and Hannan–Quinn Information Criterion (HQ).

Our estimation uses the two lags suggested by the SIC. The results of this VAR estimation are provided in Table 3. The results show that the US bond yield does not affect the FII inflows and the Indian stock market performance. The robustness of the results is further checked using other lags, such as two, six, and seven lags. We find similar relationships.

Table 3.Vector Autoregression Model
Lag 1 Lag 2 Lag 6 Lag 7
Coef. P>z Coef. P>z Coef. P>z Coef. P>z
NSE
NSE L1 -.001797 0.89 0.01 0.88 0.03 0.07 0.02 0.82
L2 0 0.99 0.01 0.07 -0.02 0.82
L3 0.04 0.08 0.02 0.76
L4 0.02 0.08 -0.03 0.68
L5 -0.05 0.08 -0.1 0.2
L6 0.05 0.08 0.02 0.76
L7 0.05 0.58
FII_EQ L1. -.001797 0.83 0 0.84 0 0.01 0.01 0.63
L2. 0 0.6 -0.01 0.01 0 0.69
L3 0.01 0.01 0.02 0.17
L4 -0.02 0.02 -0.01 0.7
L5 0 0.02 0.02 0.28
L6 -0.01 0.01 0.02 0.32
L7 0.03 0.024**
BONDY L1. -113.347 0.28 -118.04 0.26 -162.98 110.04 -127.61 0.25
L2. 52.03 0.62 117.63 110.57 127.99 0.24
L3 -35.85 112.58 -15.52 0.89
L4 171.66 112.24 161.35 0.15
L5 -198.62 110.15 -182.59 0.1
L6 24.07 107.92 22.09 0.84
L7 104.49 0.33
_cons 61.22579 0.02 62.22 0.02 61.24 28.59 65.48 0.02
FII_EQ
NSE L1. -.148478 0.81 -0.18 0.74 0.37 0.52 0.4 0.44
L2. 0.4 0.48 0.61 0.52 0.65 0.22
L3 0.41 0.55 0.47 0.39
L4 -0.17 0.56 -0.04 0.94
L5 -2.19 0.55 -2.05 0
L6 1.7 0.57 1.76 0
L7 -0.06 0.91
FII_EQ. L1. -.23241 0 -0.27 0 -0.4 0.08 -0.41 0
L2. -0.56 0 -0.71 0.08 -0.74 0
L3 -0.29 0.09 -0.32 0
L4 -0.35 0.11 -0.38 0
L5 -0.22 0.12 -0.27 0.034**
L6 -0.49 0.1 -0.55 0
L7 -0.08 0.45
BONDY L1. -1104.17 0.23 -1168.23 0.16 -1551.1 772.87 -1681.9 0.034**
L2. -718.57 0.38 -215.04 776.59 -226.86 0.77
L3 -452.48 790.7 -545.65 0.49
L4 1185.03 788.31 1290.68 0.11
L5 -1252.54 773.64 -1404.73 0.08
L6 295.83 757.94 346.73 0.66
L7 -615.48 0.42
_cons -13.29315 0.95 -51.08 0.81 -1.43 200.79 25.96 0.9
BONDY
NSE L1. .0000977 0.031** 0 0.038** 0 0 0 0.12
L2. 0 0.16 0 0 0 0.08
L3 0 0 0 0.93
L4 0 0 0 0.54
L5 0 0 0 0.3
L6 0 0 0 0.028**
L7 0 0.61
FII_EQ. L1. 0 0.24 0 0.22 0 6.87e 0 0.78
L2. 0 0.69 0 6.68e 0 0.64
L3 0 7.90e 0 0.65
L4 0 9.34e 0 0.44
L5 0 9.93e 0 0.23
L6 0 8.55e 0 0.09
L7 0 0.99
BONDY L1. 0.0354344 0.59 0.04 0.54 0.05 0.07 0.03 0.62
L2. -0.16 0.011** -0.19 0.07 -0.2 0.00***
L3 0.13 0.07 0.13 0.06
L4 -0.08 0.07 -0.08 0.23
L5 -0.08 0.07 -0.08 0.24
L6 -0.15 0.06 -0.15 0.019**
L7 -0.07 0.3
_cons -0.020469 0.21 -0.03 0.08 0.04 0.02 0.04 0.03

This table presents results obtained from the VAR model. NSE represents the Indian stock market; FII_EQ refers to the FII inflows to equity market, and the BONDY refers to the US Bond Yield. *, **, and *** denotes statistical significance at 10%, 5%, and 1%, levels, respectively.

D. Granger Causality Test

The Granger causality test is widely used to identify long-run relationships among variables in the literature. The results of the Granger causality test are provided in Table 4. Since the p-values under the Granger causality test are not significant, there is no long-run causality among the US bond yield, FII inflows, and Indian stock market performance.

Table 4.Granger Causality Wald test
Equation Excluded chi2 df Prob > hi2
Lag 1
NSE FII_EQ. 0.04503 1 0.832
NSE BONDY 1.1931 1 0.275
NSE All 1.2312 2 0.54
FII_EQ NSE NSE 0.05505 1 0.814
FII_EQ BONDY 1.4392 1 0.23
FII_EQ All 1.6313 2 0.442
BONDY NSE 4.6796 1 0.031
BONDY FII_EQ. 1.3623 1 0.243
BONDY All 4.8385 2 0.089
Lag 2
NSE FII_EQ. 0.30793 2 0.857
NSE BONDY 1.4565 2 0.483
NSE All 1.7166 4 0.788
FII_EQ. NSE NSE 0.58577 2 0.746
FII_EQ. BONDY 2.8966 2 0.235
FII_EQ. All 3.2985 4 0.509
BONDY NSE 6.439 2 0.04
BONDY FII_EQ. 1.668 2 0.434
BONDY All 8.0346 4 0.09
Lag 6
NSE FII_EQ. 6.9146 6 0.329
NSE BONDY 3.6851 6 0.719
NSE All 10.165 12 0.601
FII_EQ. NSE NSE 12.943 6 0.044
FII_EQ. BONDY 7.6136 6 0.268
FII_EQ. All 18.314 12 0.106
BONDY NSE 29.554 6 0.000
BONDY FII_EQ. 8.3529 6 0.213
BONDY All 36.049 12 0.000
Lag 7
NSE FII_EQ. 7.978 7 0.335
NSE BONDY 7.8375 7 0.347
NSE All 17.034 14 0.254
FII_EQ. NSE NSE 11.845 7 0.106
FII_EQ. BONDY 7.5955 7 0.370
FII_EQ. All 18.685 14 0.177
BONDY NSE 28.651 7 0.000
BONDY FII_EQ. 9.0254 7 0.251
BONDY All 36.597 14 0.001

This table presents the results obtained from the Granger Causality Wald test. NSE represents the Indian stock market; FII_EQ. refers to the FII inflows to equity market; and the BONDY refers to the US Bond Yield.

IV. Conclusion and Discussion

We investigate the extent to which the US bond yield affects FII inflows to India and the performance of the Indian stock market using monthly data from 2002 to 2021. Using the VAR model and the Granger causality test, we find no short- and long-run relationships among the US bond yield and FII inflows to the Indian stock market and its performance. Therefore, the US monetary policy does not affect FII inflows to the Indian stock market and its performance. This is contrary to the popular belief that “when the US (Federal Reserve) sneezes, the world catches a cold”. The role of factors such as domestic institutional investor participation and the changes in foreign exchange rate can be explored by future research.


Acknowledgement

Authors would like to thank an anonymous referee and the editorial team of Asian Economics Letters for their valuable comments which improved the quality of this paper.