Identifying Inflation Dynamics in India in the Post Reform Period

Abstract

The present study aims at examining the inflation dynamics in Indian context with a particular focus on its determinants from 1991–1992Q1 to 2017–2018Q4. The purpose of this study is to investigate the role of monetary, fiscal, structural and external variables in explaining inflationary tendencies in India in the post economic reform period. To identify the determinants fuelling the inflationary tendencies, the study employs ARDL bounds testing procedure followed by the VECM Granger causality test. The findings indicate that interest rate shock and output growth mitigates inflation while rupee depreciation, money supply generate inflationary pressures in the economy. Moreover, fiscal deficit has inflationary impact only in the short run. The positive link between inflation and openness refutes the applicability of Romer’s hypothesis in the Indian context. VECM based Granger causality indicates that money supply and interest rate causes both output and inflation, which suggests monetary policy in India has an important role to play in the process of economic growth and price stability.

JEL Classification: E3, E4, F6, E620

Keywords

Inflation money interest rate output exchange rate fiscal deficit trade openness ARDL Granger causality VECM test

Introduction

Low inflation is considered to be natural for any economy and acts as a greasing factor for economic growth, while high inflation is widely believed to impede economic growth and causes economic uncertainty. Therefore, price stability, as defined by a low and stable inflation is a prerequisite for financial stability and economic growth in all the countries. It is a well-known fact that inflation has been the prime concern for India’s monetary authorities. To ensure low and stable inflation, the monetary policy framework in India has undergone a significant transformation from being credit regulatory authority up to the mid-1980s to recently adopted inflation targeting approach in 2015 via monetary targeting approach in 1985 and multiple indicator approach in 1998. Stabilizing inflation with a reasonable economic growth is always a challenging task, particularly for an emerging economy like India. It is an issue that harms the real economy, especially when a large part of the population has no hedge against inflation. Recently, the issue has attracted lots of attention when the monetary policy framework of the Reserve Bank of India (RBI) has shifted to inflation targeting from multiple indicator approach. Under this approach, the prime objective of monetary policy is to ensure low and stable inflation of 4% while supporting output growth.

The problem of inflation has been with mankind ever since we moved from barter system to the use of the medium of exchange, such as paper currency, precious metals or even cigarettes, as happened in a prisoner of war camp during the Second World War (Radford, 1945). Before the Second World War, inflation was mainly in response to some particular events such as war, gold discovery or unmanageable economic crises and they continued as long as the event with which they were associated lasted. However, post-Second World War, the western industrial World has witnessed an almost unbroken rise in prices and money income. Inflation, in fact, has been a general phenomenon from the 1960s to 1980s and both the developed as well as underdeveloped economies experienced a very high rate of inflation. The rate of inflation has been higher in developed countries from 1960 to 1970. However, from the 1970s to 1980s it was developing countries that faced high inflation, mainly attributed to supply shock in the form of two oil price shocks. The inflation period from 1965 to 1984 came to be known as ‘great inflation’ (Meltzer, 2004). Great inflation has also been one of the root causes for the collapse of the Bretton Woods system of fixed exchange rate in 1971 and the separation of the US dollar from its last link to gold.

The world’s biggest episodes of high inflation occurred in the nineteenth century in Europe once in 1923 in Germany and again around 1946 in Hungary. The post-Second World War hyperinflation of Hungary held the record.¹ The German hyperinflation of 1923 may well be the most analysed and diagnosed inflation. It played havoc with the economy, created political tensions that contributed to the rise of Nazism, and also caused psychological disturbances. Doctors in Germany in 1923 identified a mental illness called ‘cipher stroke’ that affected many people during the height of hyperinflation (Basu, 2011). More recently, hyperinflation occurred in Zimbabwe in 2008 when prices doubled every 24.7 hours, and the country finally decided to stop printing its currency and currencies from other countries were being used.

Like other economies, India too witnessed episodes of high inflation during the 1970s and 1980s reflecting a mix of expansionary fiscal policy, accommodating monetary policy and supply shocks. High inflation during the crisis year of 1991–1992 on account of the balance of payment (BOP)² problems and high fiscal deficit (8.4% of GDP) was one of the major reasons that led to 1991 economic reform initiative. But despite the economic reforms, the economy has not been in a position to overcome the menace of the inflationary situation. After GFC inflation in India remained near double-digits.³ Even more recently, Indian economy witnessed spike in inflation in the second half of 2019 which got further aggravated following the COVID-19 Pandemic.⁴ Thus, inflation continued to impose significant challenges for policy makers in India also.

A Brief Sojourn of Inflationary Trends in India Since Independence

Table 1 provides the average decadal inflation since 1951–1952. The long-term inflation trend in India from 1951–1952 to 2020–2021 (5.98%) indicates that historically, inflation in India has remained moderate. Moreover, the average inflation in the post reform period is 5.47%. From Table 1 and Figure 1, long run inflation trend in India suggests that inflation was highly volatile in the first three decades of the 1950s, 1960s and 1970s. However, since mid-1980s, the inflation has moderated in the subsequent decades with some occasional spikes in inflation. It has remarkable stability coinciding with the great moderation⁵ globally except for the high inflation episode during 2008–2009, 2010–2011 and 2011–2012.

Table 1.

Decadal WPI Inflation Rate in India from 1951–1952 to 2018–2019 (%)

Decade	Average Inflation	Maximum Inflation	Minimum Inflation	Standard inflation
1951–1952 to 1960–1961	1.90	14.00	–12.5	7.72
1961–1962 to 1970–1971	6.20	13.90	–1.10	4.89
1971–1972 to 1980–1981	10.30	25.20	–1.10	9.31
1981–1982 to 1990–1991	7.20	10.30	4.40	3.84
1991–1992 to 2000–2001	7.80	13.70	3.30	3.48
2001–2002 to 2010–2011	5.50	9.60	3.40	2.08
2011–2012 to 2020–2021	3.05	8.90	–3.70	3.51
Long-term trend 1951–1952 to 2020–2021	5.98
1991–1992 to 2020–2021	5.47

Source: Handbook of Statistics on Indian Economy, RBI and Office of the Economic Advisor, GOI.

Figure 1.

Source: Handbook of Statistics on Indian Economy, RBI, various issues, and Office of the Economic Advisor, GOI official site.

The decadal average inflation figures and falling standard deviation in Table 1 since the 1950s reveals that both inflation and its volatility has declined over a period of time. One of the factors attributed to this could be the effectiveness of monetary policy in reducing inflation and its volatility. The fall in inflation and its volatility since 1990s is also attributed to success of economic reforms of the early 1990s.

Review of Empirical Literature

The issue of inflation has always been at the core of theoretical and empirical debates. Researchers have examined the inflationary trends in various countries with various determinants by applying different sample sizes using different methods. Jadhav and Singh (1990) has examined the impact of budget deficit, money supply, on inflation for the sample period 1970–1971 to 1987–1988 in India. The study found that an increase in the price level produces more budget deficits, which ultimately increased the money supply and this increased money supply, in turn, leads to further inflation. The study concludes that there is a self-perpetuating process of deficit-induced inflation and inflation-induced deficits in the Indian context. Based on quarterly dataset from 1982–1998Q2, Challen and Chang (1999) by developing a series of bivariate vector autoregressive (VAR) models states that broad money and narrow money have much impact on the future development of inflationary pressures. Besides, import prices, exchange rate, stock prices and the prices of primary articles also provide useful information about movements in future price level.

Parida et al. (2001) studies the relationship between fiscal deficit, money supply and price level in the Indian context for the sample period 1960–1961 to 1999–2000. The study concludes that fiscal deficit is an essential driver of inflation in India via monetized deficit, which leads to an increase in money supply and thus causes inflation. Das (2003), by applying VARMA technique the study indicates that there exists a bidirectional causality between money and inflation. Likewise, output also affects the price level and vice-versa. By employing VAR model, John (2003) investigated the determinants of inflation in India over the period January 1991 to December 2001. The study finds that growth in monetary aggregates and nominal effective exchange rates causes inflation.

Ashra et al. (2004) through the application of Engle-Granger cointegration test shows that there exists bidirectional causality between inflation and money supply. The study also did not find any linkage between government deficit and money supply and consequently, the study argues that fiscal deficit is no more relevant for price stability in India. Khundrakpam (2008) tested the presence of exchange rate pass-through (ERPT) to domestic prices in India for the post reform period. The findings revealed that the exchange rate has a positive impact on inflation. Moreover, the study shows that Money supply, output and persistence of inflation is an essential factor for the rise in price level. Khundrakpam and Pattnaik (2010) explored the linkages between fiscal deficit and inflation over the pre fiscal responsibility and budget management (FRBM) period 1953–2005 as well as during the period of 1953–2009. The study reveals that fiscal deficit is inflationary. RBI (2012), also reported evidence in favour of the fiscal inflation in India. Thus that targeting fiscal deficit as a tool for stabilization continues to remain valid. Recently, Nguyen (2015) also reported the evidence of fiscal INF in case of eight selected economies of Asia. More recently, Ramu and Gayithri (2017) through SVAR approach have identified some transmission channels by which fiscal deficit impacts inflation in India.

Khundrakpam and Pattanaik (2010) by applying cointegration method assess the importance of both demand and supply side factors in inflation. The study concludes that Indian inflation is mainly caused by demand-side factors, while supply-side factors in the form of import prices also affect inflation but only in the short run. Similarly, Mishra and Mishra (2010) investigated both the domestic and external determinants of inflation in India by using VAR model for the monthly data 1996 to 2005. The results show that volatility in Indian inflation in the post reform era is mainly caused by global supply factors and external fluctuations. Moreover, the study also revealed a larger and quicker impact of global supply shocks and external shocks on domestic inflation.

Lin and Chu (2013) empirically examined the relationship between fiscal deficit and inflation for 91 countries by using annual data from 1960 to 2006. The findings of the article reveal that fiscal deficit has a substantial impact on inflation in high inflation episodes because of faster money creation during high inflationary periods and low impact in low inflation episodes. Thus, the study suggests that fiscal consolidation would be more effective in controlling price movements when the inflation rate is high. Jalil et al. (2014) tested the fiscal theory of inflation in the case of Pakistan from 1972 to 2012. The ARDL estimation results show fiscal deficit generate inflationary pressures in Pakistan.

Mohanty and Bhanumurthy (2014) by using monthly data did not find evidence of exchange rate pass through. The authors argue that this could be as a result of offsetting the sterilisation policy undertaken by RBI during expansionary money supply growth. Mohanty and John (2015) applied the time-varying SVAR technique to analyse the impact of various determinants of inflation in India. The article highlights the role of monetary and fiscal policy in the containment of inflation irrespective of the nature of the shock to the inflation process. Maitra (2016) investigated the relative role of structural as well as monetary factors in explaining CPI inflation in India in a VAR framework. The study holds that the role of output gap via demand pressures in explaining price level is more significant than the money supply, interest rate and exchange rate. Bhat and Sharma (2020) in an asymmetric procedure shows that output growth and interest rate has inflation dampening effect which fiscal deficit and openness has inflationary impact in the Indian context over the period 1970–2016.

Romer (1993) argues that economies that are more open tends to have low rate of inflation. Lane (1997) have explained the view why open economies have lower inflation. Rogoff (1992) model shows that open economies gain less from surprise inflation and pay the price of monetary expansion more quickly, particularly if the country is following market determined exchange rate. While developing a theoretical model, Cooke (2010) justifies that inflation depends on openness and a more open economy tends to have lower inflation. In Indian context, Joshi and Debashish (2010) investigated the relationship between openness and inflation using quarterly data from 1984–1985 to 2004–2005. The study concluded that trade openness have significantly contributed to the disinflation process, and the relationship has grown strong after the liberalization period. However, more recently, Ajaz et al. (2016) by using annual time series data from 1970–2014 refutes the Romer’s (1993) in the Indian context. Thus, the study argues that gains from openness do not accrue without the cost in terms of inflation and its volatility.

The above literature survey highlighted certain important points. First, the studies with respect to annual data fails to capture short run inflation dynamics appropriately which reduces their relevance for policy purposes. Moreover, annual studies span as many as three to four decades must take into account the various structural breaks that the any economy might have undergone over the years. Unfortunately, this treatment is also missing in most studies. Second, overall, the past studies have investigated some aspects of the determinants of inflation, but a comprehensive assessment considering all potential determinants (monetary variables, fiscal variable and external variable) is missing. Mohanty and John (2015) argues that though the inflationary pressures in India have been theoretically ascribed to both domestic and foreign factors and to both supply and demand shocks, however, the empirical evidence remains inadequate and provide mixed results. The possible reason for such an inadequacy could be the changes in determinants of inflation over the period of time. Thus, the study adopts a broader framework by taking all potential macroeconomic determinants provoking inflation. Further, the inclusion of so many determinants or variables provides consideration to issues of model adequacy, such as the possibility of omitted variables bias. The present study is exclusively done for the post economic reforms period, which is based on a large sample size starting from 1991–1992Q1 to 2017–2018Q4.

Database and Research Methodology

Dataset

The study employs quarterly data on a financial year basis from 1991–1992Q1 to 2017–2018Q4. The data for all variables have been collected from the Handbook of Statistics on Indian Economy published by the RBI and Office of the Economic Advisor, Government of India (GOI). WPI is used as a measure of inflation. As the determinants of inflation, the study involves money supply, interest rate, real GDP, fiscal deficit, exchange rate and finally, trade openness. The quarterly series of inflation, money supply, interest rate, exchange rate, fiscal deficit and trade openness are not readily available in the RBI databank and hence, the data are extracted from the monthly series of variables concerned. With due care, all monthly series are converted into quarterly series. Importantly, the concept of stock and flow⁶ of the variables have also been taken into account while converting the dataset from monthly to quarterly series. The published quarterly time series data on real Gross Domestic Product (GDP) data is available only from 1996–1997Q1. Hence, quarterly data for the earlier period is obtained from annual data on real GDP by using an optimal interpolation procedure.⁷ Likewise, monthly data on the exchange rate is available from 1992–1993. Hence, quarterly data for the earlier period is obtained from the annual exchange rate of the rupee.

Franses and Franses (1998) states that it is common to believe that seasonality must be there in case of the monthly or quarterly dataset. The presence of seasonality in a time series can add to the overall volatility of the series. Hence keeping in mind the seasonality issue, seasonality was observed in WPI, GDP and money supply. Therefore, this study utilizes seasonally adjusted quarterly data of WPI, real GDP and money supply. The seasonal adjustment was made by using the recent methodology X-13 Autoregressive Integrated Moving Average (ARIMA) SEATS of the US Census Bureau (2013).⁸ Further, in order to maintain continuity in the time series data, each series is made consistent by adjusting for a common base-year 2004–2005. Splicing method and Linking factor is employed to combine index numbers with different base-years into a single series. The base-year was changed to 2011–2012 from 2004–2005 in May 2017. The change in base-year was accompanied by changes in commodity basket of wholesale price index (WPI), including weightage. There were also methodological changes in the calculation of GDP. Hence, the study uses the old series with base-year 2004–2005. Finally, all the variables except interest rate (since the rate of interest is in rate form) have been transformed into a natural logarithmic (ln) form to reduce the problem of heteroscedasticity. Das (2003) states that logarithmic transformation makes series near symmetric and homoscedastic.

The WPI based inflation is chosen for the study because, during our sample period the RBI was primarily focusing on WPI inflation in the absence of a nation-wide representative consumer price index (CPI). Moreover, the monetary policy formulation was mainly governed by WPI upto 2014 as against CPI. The study employs a weighted average overnight call money rate (CMR) as a proxy for the interest rate. This is because the RBI infers the liquidity situation in the market by observing movements in the CMR and accordingly adjusts its policy instrument. Urjit Patel Committee (2014), while recommending flexible inflation targeting regime, also advised continuing with CMR as the operating target. Moreover, Bhoi et al. (2017), also show that CMR is the most appropriate variable. This will further reflect the impact of monetary policy on inflation. For money supply the study employs M3 which is considered as the most important indicator of the money supply. Real GDP is considered to reflect output in an economy. GDP variable sounds to be more appropriate as it accounts for the actual production process occurring within a country. Further, GDP includes all the three sectors, including primary, secondary, and tertiary. GDP series has been deseasonalized after testing for the presence of seasonality by using the X-13 ARIMA filter. Hence, the study utilizes seasonally adjusted quarterly GDP. The exchange rate is measured by the nominal exchange rate of rupee in terms of the US dollar. The inclusion of this variable is intended to capture the degree of ERPT. The gross fiscal deficit as a percentage of GDP which is the combined deficit level of Central and state governments has been used to measure the impact of fiscal policy on inflation. A recent contribution to inflation theory is the development of fiscal theory of price level (FTPL)⁹ which postulates that price level is primarily determined by government debt and fiscal policy, with monetary policy playing an indirect role. Trade openness has been included in the model to verify the validity of Romer’s (1993) hypothesis in the Indian context. Trade (sum of exports and imports) as a ratio of GDP of a country is used to measure the level of trade openness.

Econometric Methodology

Unit Root Tests

While estimating any type of long run relationship (cointegration), the first step is to determine, whether the series under study is stationary or non-stationary. To carry this exercise, the study employs two well-known non-stationarity tests; Augmented-Dicky Fuller (ADF) test Phillips-Perron (PP) test.

Model Specification

From the theoretical perspective, inflation is caused by a number of monetary, fiscal, supply side and external variables. The relationship between inflation and such determinants, in a functional form can be written as:

\begin{matrix} \ln π = α_{0} + λ_{1} D 1 + β_{1} \ln M 3 + β_{2} I R + β_{3} \ln Y + β_{4} \ln F D \\ + β_{5} \ln E X R + β_{6} \ln T O P + u_{t} \end{matrix}

(1)

Where ln = log form, t = time and μ_t = error term, which is supposed to satisfy all the properties of the Classical Linear Model.

We also added a dummy variable (D1) in the model to represent and measure the impact of the global financial crisis (that hit the global economy in 2008) on inflation. D1 takes value 1 for the period after 2008–2009Q1 and 0 before 2008–2009Q1.The structural break is tested by Chow test (Chow, 1960).¹⁰ In this test, we have to specify the date of the structural break. The Chow test for structural break requires strictly exogenous regressors and a break-point specified in advance. According to the Chow test, we do not reject the null hypothesis of ‘no structural break’ if the computed F-statistic value does not exceed the critical F-statistic value from the F-table.

ARDL Bounds Testing Model to Cointegration

The ARDL model introduced initially by Pesaran and Shin (1998b) and further extended by Pesaran et al. (2001) is used to examine the long run and short run inflation dynamics between inflation and its various determinants in India. This technique is also called the Bounds test. In a comparison of other cointegration methods the ARDL bounds testing framework to cointegration enjoys certain econometric advantages (For a detailed discussion on advantages of ARDL model, see, Narayan (2004)). The only limitation of ARDL model is that this approach fails to accommodate the variables which are integrated of order second, I (2) or higher order, and thus becomes inefficient. In the absence of prior information about the long run relationship among the variables, the following Unrestricted Error Correction Regression is estimated:

\begin{matrix} ⊿ ​ \ln π_{t} = α_{0} + λ_{1} D 1 + \sum_{i = 1}^{p} β_{1} ⊿ ​ \ln π_{t - i} + \sum_{i = 0}^{p} β_{2} ⊿ ​ \ln M 3_{t - i} + \\ \sum_{i = 0}^{p} β_{3} ⊿ ​ I R_{t - i} + \sum_{i = 0}^{p} β_{4} ⊿ ​ \ln Y_{t - i} + \sum_{i = 0}^{p} β_{5} ⊿ ​ \ln F D_{t - i} + \\ \sum_{i = 0}^{p} β_{6} ⊿ ​ \ln E X R_{t - i} + \sum_{i = 0}^{p} β_{7} ⊿ ​ \ln T O P_{t - i} + δ_{1} \ln π_{t - 1} + \\ δ_{2} \ln M 3_{t - 1} + δ_{3} I R_{t - 1} + δ_{4} \ln Y_{t - 1} + δ_{5} \ln F D_{t - 1} + \\ δ_{6} \ln E X R_{t - 1} + δ_{7} \ln T O P_{t - 1} + μ_{t} \end{matrix}

(2)

Where Δ denotes the first difference operator, a₀ is the drift component, μ_t is the usual white noise residuals assumed to satisfy all the basic assumptions of the classical linear regression model, t stands for time element while the variables lnP, lnM3, IR, lnY, lnFD, lnEXR and lnTOP are same as defined earlier. The coefficient β₁ – β₇ represents the short run dynamics of the model, while coefficients δ₁ – δ₇ indicate long run inflation dynamics. λ₁ is the coefficient of the dummy variable (D1), which is supposed to measure the impact of a structural break of 2008 in the form of a global financial crisis.

To investigate the existence of long run relationship (called cointegration) among the variables of the system, the bounds test is used. The bounds test is based on the F-test under the null of no long run relationship. The null hypothesis of no cointegration among the variables in the Equation (2) is expressed as:

H_{0} : δ_{1} = δ_{2} = δ_{3} = δ_{4} = δ_{5} = δ_{6} = δ_{7} = 0 (N o c o int e g r a t i o n)

Against the alternative hypothesis:

H_{1} : δ_{1} \neq δ_{2} \neq δ_{3} \neq δ_{4} \neq δ_{5} \neq δ_{6} \neq δ_{7} \neq 0 (C o int e g r a t i o n e x i s t s)

This can be denoted as

F (\ln π / \ln M 3, I R, \ln E X R, \ln Y, \ln F D, \ln T O P)

Given that a long run relationship exists, a further two-step procedure to estimate the model is undertaken. Pesaran and Shin (1998a) show that valid asymptotic inferences on short run and long run parameters can be made under least squares estimates of an ARDL model. Having found a cointegrating relationship among specified variables, Equation 2 is estimated using the following ARDL model.

\begin{matrix} \ln π_{t} = α_{0} + λ_{1} D 1 + \sum_{i = 1}^{p} β_{1} \ln π_{t - i} + \sum_{i = 0}^{p} β_{2} \ln M 3_{t - i} + \\ \sum_{i = 0}^{p} β_{3} I R_{t - i} + \sum_{i = 0}^{p} β_{4} \ln Y_{t - i} + \sum_{i = 0}^{p} β_{5} \ln F D_{t - i} + \\ \sum_{i = 0}^{p} β_{6} \ln E X R_{t - i} + \sum_{i = 0}^{p} β_{7} \ln T O P_{t - i} + μ_{t} \end{matrix}

(3)

The lag order in the ARDL model is selected by AIC before the selected model is estimated by the ordinary least squares technique.¹¹ In the presence of cointegration, short run elasticities are calculated by applying an ECM of the following form:

\begin{matrix} ⊿ ​ \ln π_{t} = λ_{1} D 1 + \sum_{i = 1}^{p} β_{1} ⊿ ​ \ln π_{t - i} + \sum_{t = 0}^{p} β_{2} ⊿ ​ \ln M 3_{t - i} + \\ \sum_{i = 0}^{p} β_{3} ⊿ I R_{t - i} + \sum_{i = 0}^{p} β_{4} ⊿ ​ \ln Y_{t - i} + \sum_{i = 0}^{p} β_{5} ⊿ ​ \ln F D_{t - i} + \\ \sum_{i = 0}^{p} β_{6} ⊿ ​ \ln E X R_{t - i} + \sum_{i = 0}^{p} β_{7} ⊿ ​ \ln T O P_{t - i} + ψ E C M_{t - 1} + μ_{t} \end{matrix}

(4)

Where βs indicates the short run impact of explanatory variables on inflation. ECM is the lagged error correction term and ψ is the coefficient of lagged error correction term. The coefficient of ECM measures the speed of adjustment per quarter towards the long run equilibrium following a system shock in the short run which is supposed to be negative and statistically significant. The negative and statistically significant coefficient of ECM further confirms the long run relationship among the variables. The coefficient of lagged error term, is also known as an adjustment parameter.

Finally, as suggested by Pesaran and Pesaran (1997), the stability of the short run and long run coefficients is tested through the cumulative sum of recursive residuals (CUSUM) and cumulative sum of squares of recursive residuals (CUSUMSQ) tests proposed by Brown et al. (1975).

Granger Causality Test

After analysing the presence of cointegrating association among variables, we apply the vector error correction model (VECM) Granger causality test to check the direction of causation between the variables. The VECM estimate is given as follows:

[\begin{matrix} Δ ​ \ln π ​_{t} \\ Δ ​ \ln π M 3 ​_{t} \\ Δ I R ​_{t} \\ Δ ​ \ln Y ​_{t} \\ Δ ​ \ln F D ​_{t} \\ Δ ​ \ln E X R ​_{t} \\ Δ ​ \ln T O P ​_{t} \end{matrix}] = [\begin{matrix} β_{1} \\ β_{2} \\ β_{3} \\ β_{4} \\ β_{5} \\ β_{6} \\ β_{7} \end{matrix}] + \sum_{i = 1}^{q} (1 - B) [\begin{matrix} c_{11 i} c_{12 i} c_{13 i} c_{14 i} c_{15 i} c_{16 i} c_{17 i} \\ c_{21 i} c_{22 i} c_{23 i} c_{24 i} c_{25 i} c_{26 i} c_{27 i} \\ c_{31 i} c_{32 i} c_{33 i} c_{34 i} c_{35 i} c_{36 i} c_{37 i} \\ c_{41 i} c_{42 i} c_{43 i} c_{44 i} c_{45 i} c_{46 i} c_{47 i} \\ c_{51 i} c_{52 i} c_{53 i} c_{54 i} c_{55 i} c_{56 i} c_{57 i} \\ c_{61 i} c_{62 i} c_{63 i} c_{64 i} c_{65 i} c_{66 i} c_{67 i} \\ c_{71 i} c_{72 i} c_{73 i} c_{74 i} c_{75 i} c_{76 i} c_{77 i} \end{matrix}] \times [\begin{matrix} Δ ​ \ln π ​_{t - 1} \\ Δ ​ \ln M 3 ​_{t - 1} \\ Δ I R ​_{t - 1} \\ Δ ​ \ln Y ​_{t - 1} \\ Δ ​ \ln F D ​_{t - 1} \\ Δ ​ \ln E X R ​_{t - 1} \\ Δ ​ \ln T O P ​_{t - 1} \end{matrix}] + [\begin{matrix} λ_{1} \\ λ_{2} \\ λ_{3} \\ λ_{4} \\ λ_{5} \\ λ_{6} \\ λ_{7} \end{matrix}] E C T_{t - 1} + [\begin{matrix} μ_{1 t} \\ μ_{2 t} \\ μ_{3 t} \\ μ_{4 t} \\ μ_{5 t} \\ μ_{6 t} \\ μ_{7 t} \end{matrix}]

Where 1-B is the lag operator, ECT_t_–1 is the lagged error correction term. λ₁–λ₇ are the coefficients of the lagged error correction terms. μ_ts are serially independent random errors with mean zero and finite covariance matrix. The long run causality is confirmed by the negative and the statistically significant coefficients of lagged error correction terms by applying a t-test. While as, short run causality is determined by the joint significance of the lagged explanatory values by apply Wald test (Wald test is based on F-statistics or Chi-square statistics).

Empirical Results and Discussion

Unit Root Test Results

The results of PP non-stationarity test under the null hypothesis of (Series is non-stationary) in Table 2 shows that none of the variables is integrated of order 2.¹²

Table 2.

Results of Phillips-Perron Test

	Null Hypothesis: The Series has Unit Root
	Only Intercept		Intercept and Trend Both
Series	PP value	p-value	PP value	p-value	Decision
Inπ	–1.398(3)	.579	–0.769(2)	.964	Non-stationary
IR	–5.207*(3)	.000	–5.171*(3)	.000	Stationary
lnM3	–1.863(17)	.348	–0.386(17)	.986	Non-stationary
lnY	1.009(7)	.996	2.841(3)	.186	Non-stationary
lnFD	–7.008*(1)	.000	–7.863*(3)	.000	Stationary
lnEXR	–1.213(3)	.665	1.944(3)	.622	Non-stationary
lnTOP	–1.229(7)	.658	–1.101(6)	.936	Non-stationary
∆(lnπ)	–7.014*(9)	.000	–7.627*(10)	.000	Stationary
∆(IR)	–21.611*(4)	.000	–21.32*(20)	.000	Stationary
∆(lnM3)	–12.598*(8)	.000	–14.103*(3)	.000	Stationary
∆(lnY)	–9.926*(8)	.000	–9.959*(8)	.000	Stationary
∆(lnFD)	–21.477*(22)	.000	–22.05*(22)	.000	Stationary
∆(lnEXR)	–6.945*(1)	.000	–6.905*(1)	.000	Stationary
∆(lnTOP)	–7.901*(12)	.000	–7.912*(14)	.000	Stationary

Source: The author.

Note: Parenthesis contain lags. The Bartlett Kernel with Newey–West automatic bandwidth selection criterion has been used. ∆ represents first difference of a time series. *Indicates rejection of null at 1% level of significance.

The mixed order of integration further justifies the use of the ARDL procedure for the estimation of a long run and short run relationship between inflation and explanatory variables in the present study. Before we proceed to cointegration test, the study applies the Chow (1960) structural break test with a known break date to check whether the global financial crisis has any impact on inflation. The finding of the Chow test confirms the presence of a structural break in the WPI series at a specified breakpoint, that is, 2008–2009Q1 with F-statistics value 31.49 (p-value of computed F-statistic is less than 5%). This was the period during which the world was hit by the North Atlantic financial crisis, which later came to be known as the global financial crisis. To account for the structural break, a dummy variable (D) is added in the ARDL model to avoid model specification error.

Cointegration Test Results

To carry out the bounds tests, the equation is estimated following the OLS procedure and F-statistics is computed under Wald test on the joint hypothesis that the coefficients of the lagged level variables are jointly equal to zero. Before proceeding to the bounds test for cointegration, the lag length is selected on the basis of AIC. The study will choose a maximum lag length of 4 for both endogenous and exogenous variables as the study is utilizing quarterly data.

The results of the bounds test is reported in Table 3. The calculated F-statistic value (6.450) is greater than the upper bound critical value at 1% level of significance when inflation is dependent variable and money supply, interest rate, real output, fiscal deficit, exchange rate and trade openness are used as explanatory variables. The results clearly indicate the strong evidence of existence of a long run relationship (cointegration) between inflation and the explanatory variables for the sample period 1991–1992Q1 to 2017–2018Q4. In other words, this implies that money supply, interest rate, real output, fiscal deficit, exchange rate and trade openness has a long run impact on inflation.

Table 3.

Bounds Test Results for Cointegration

Model = f(Inπ/InM3, IR, InFD, InEXR, InTOP)

	Null Hypothesis: No Cointegration Exists
Level of Significance	Lower Bound	Upper Bound	Wald Test (F-statistic)	Decision
1%	3.06	4.15
5%	2.39	3.38	6.450*	Cointegration
10%	2.08	3.00

Source: The author.

Note: * denotes statistical significance at 1% level.

Analysis of the Long Run Coefficients of ARDL

The long run ARDL findings in Table 4 indicate that except fiscal deficit all the coefficients are statistically significant and have signs as expected and supported by the economic theory. Money supply, exchange rate and trade openness have a positive impact on inflation while interest rate and real output have a negative impact on inflation in India. However, fiscal deficit turns out to be statistically insignificant but the positive coefficient of fiscal deficit is in line with the fiscal theory of inflation. The results are discussed in detail below.

Table 4.

Estimated Long Run ARDL Model

Dependent variable = lnπ (WPI)

Selected model (2, 1, 1, 1, 1, 2, 1) based on AIC

Variable	Coefficient	Standard Error	t-statistics	p-value
Constant	4.287**	2.183	1.963	.053
lnM3	0.474*	0.165	2.876	.005
IR	–0.186**	0.103	–1.808	.074
lnY	–0.709*	0.081	–8.733	.000
lnFD	0.219	0.157	1.391	.168
lnEXR	0.593**	0.302	1.961	.053
lnTOP	0.465*	0.054	8.560	.000
D1	0.081**	0.041	1.962	.053
R² = 0.977 DW statistics = 1.952 F-statistics = 6,112.844 (0.000)

Source: The author.

Note: * and ** denotes statistical significance at 1% and 10% level, respectively.

First, the coefficient of the money supply comes out to be positive and statistically significant at 1% level. This demonstrates that keeping other things constant, 1% growth in money supply (M3) in an economy will lead to 47.5 basis points rise in the price level. This result also supports the Monetarists view, which postulates that any rise in money supply will cause the price level to go up. Therefore, Friedman’s view that inflation is a monetary phenomenon holds in case of India. The result seems logical in case of developing economies like India which are constrained with supply-side bottlenecks, lack of technological advancements and resource mobilisation (Maitra, 2016). Therefore, an increase in money supply may not raise real output but can bring pressure on prices. It is important to mention here that with the economic liberalisation of 1990s, an increase in money supply was mostly due to a rise in capital inflows or foreign reserves.¹³ This rise in the money supply might have caused inflation in the post reform period. This result is in line with the findings of Khundrakpam and Pattanaik (2010) who also found that growth in money supply plays an important role in causing variations in inflation in the long run. However, this finding contradicts the results of Sabade (2014) who could not establish any clear relation between money and inflation in India. The difference in results might have raised on account of different sample sizes. Sabade’s study is based on a small sample size covering the period 2010–2013.

Second, the coefficient of interest rate is negative (–0.186) and statistically significant, although weakly, at the 10% level.¹⁴ This negative coefficient of interest rate implies that positive interest rate shock or innovation dampens inflation. Thus, keeping other factors constant, 1% rise in interest rate will bring down price level by 18.6 basis points. Based on this finding, we are in a position to conclude that the contractionary monetary policy by the RBI has a significant impact on mitigating inflation in India. The negative impact of interest rate on inflation is in conformity with the macroeconomic theory that higher is the lending rate of interest, higher is the cost of capital. This higher cost of capital causes investment spending and aggregate demand to fall which may lead to lower economic growth as well as inflation. Also, higher interest rates reduces demand for money and thus does not assert inflationary pressure in the economy. This implies that a positive interest rate shock can be used by the RBI as an anti-inflationary policy instrument? Mohanty and John (2015), Khundrakpam (2008) and Kapur and Behera (2012) also observed that an increase in the nominal rate of interest has a negative impact on inflation in India.

Third, the coefficient of output is also negative and statistically significant which indicates that any increase in output has the tendency to reduce the inflation level. The negative coefficient of real output states that supply-side factors play a crucial role in controlling inflation in India, that is, as and when the supply of output increases, it has the tendency to lower the level of prices. The negative coefficient (–0.709) of real output indicates that 1% growth in the supply of output, ceteris paribus would decrease inflation by 70.9 basis points. This result is consistent with the belief of Structuralist economist’s that supply-side bottlenecks mainly cause inflation in developing economies like India. Thus, targeting supply-side constraints will help to moderate inflation. This finding is consistent with the findings of Yadav and Lagash (2011) and Nair (2014), who, while using IIP as a proxy for output, found a negative impact of output on inflation in the post reform period.

Fourth, the coefficient of fiscal deficit turns out to be positive but statistically insignificant. The positive coefficient is in line with the fiscal theory of price level. This implies that fiscal deficit has not provoked inflation in the post reform period. Therefore the two channels through which fiscal deficit could cause inflation, that is, by exerting pressure on aggregate demand in relation to potential output and by leading to an excessive expansion in money growth were almost absent in the sample period. This may be due to the historic accord between the RBI and Central government for phasing out the issue of ad hoc treasury bills, thereby eliminating automatic monetisation of a budget deficit. Further, the enactment of FRBMA Act, 2003, by the Central government leads to a reduction in fiscal deficit,¹⁵ which further enhanced the autonomy of monetary authorities to achieve the objective of price stability. This result is in line with the findings of Balakrishnan (1997) and Kaur (2021), who also observes that fiscal deficit is not provoking inflation in India. However, this finding is against the finding of Khundrakpam and Pattanaik (2010), who found a significant positive impact of fiscal deficit on inflation. The difference in results could arise due to the reason that the authors had used annual data spanning for the period 1953 to 2009.¹⁶

Fifth, the coefficient of the exchange rate is found to be positive with a value of 0.593 and is statistically significant at 1% level of significance. This positive coefficient of exchange rate shows that a 1% rise in the exchange rate of rupee against a dollar (depreciation) will increase inflation in India by 59.3 basis points.¹⁷ The positive coefficient of an exchange rate with respect to inflation by less than 1% further indicates that the ERPT to domestic prices in India is incomplete.¹⁸ The positive association between inflation and exchange rate is consistent with the conventional purchasing power parity.¹⁹ Bhattacharya et al. (2008), Khundrakpam (2008) and Kapur and Behera (2012) and Saha and Zhang (2013) also found the presence of ERPT to domestic prices in the Indian context.

The estimated coefficient of openness is 0.465 and is statistically significant at 1% level of significance. According to the estimation, 1% rise in trade openness leads to a 46.5 basis points rise in inflation in India. Thus, this finding also refutes the Romer’s (1993) hypothesis that open economies tend to have lower inflation and concludes that openness has resulted in higher prices in India in the post reform period. The positive linkage between inflation and openness found in our study is consistent with the general notion that globalisation is inflationary for developing economies, as in India (Evans, 2007; Jalil et al., 2014). The positive impact of trade openness on inflation might be because of the importance of imports, particularly oil and other manufactured products in trade basket, which has a significant impact on Indian inflation because of increasing oil and manufactured prices at the global market. However, at times global oil prices have decreased as in 2014 when global oil prices fell by more than 50%, but the pass-through has been either absent or very limited in case of India. This may be due to the fact that energy prices are administered in India.

Lastly, the coefficient of a dummy variable turned out to be positive and statistically significant meaning thereby the global financial crisis has an adverse impact on India in the form of high inflation. This finding seems logical because India embarked on a large fiscal and monetary stimulus in the post-crisis period (Singh, 2012).

Short Run Inflation Dynamics

The results of short run inflation dynamics is reported in Table 5. In the short run all variables including fiscal deficit are statistically significant and have expected coefficients in line with theoretical predictions. The coefficient on the lagged error correction term (ECM_t_–1) is negative and statistically significant at 1% level, which means that the inflation series is non-explosive and equilibrium in the long run is attainable. This result also reinforces the earlier cointegration results under bounds test between inflation and explanatory variables. The negative coefficient of the lagged error term (–0.343) suggests that long run equilibrium is stable²⁰ and the speed of adjustment to restore the long run equilibrium is 34% per quarter following a shock in the short run. This also means that in the long run all variables move simultaneously towards equilibrium.

Table 5.

Error Correction Representation for the Selected ARDL Model

Variable	Dependent Variable= lnπ
Variable	Coefficient	Std-error	t-statistics	p-value
∆(In π_t–1)	0.327*	0.101	3.217	.002
∆(In π_–2)	–0.013	0.009	–1.312	.196
∆(In M3_t)	0.134**	0.063	2.112	.038
∆(In M3_t–1)	0.171*	0.062	2.747	.008
∆(IR_t)	–0.094***	0.049	–1.890	.065
∆(IR_t–1)	–0.081***	0.048	–1.687	.099
∆(In Y_t)	–0.272*	0.072	–3.783	.000
∆(In Y_t–1)	–0.220**	0.091	–2.397	.021
∆(In FD_t)	0.037*	0.011	3.238	.002
∆(In FD_t–1)	0.115	0.115	1.004	.320
∆(In EXR_t)	0.157*	0.045	3.448	.001
∆(In EXR_t–1)	–0.014	0.050	–0.287	0.774
∆(In EXR_t–2)	0.035	0.026	1.338	0.187
∆(In TOP_t)	0.096*	0.015	6.199	0.000
∆ (In TOP_t–1)	0.030***	0.016	1.902	0.061
D1	0.030*	0.007	7.796	0.000
ECM_t–1	–0.342*	0.050	–6.840	0.000
Regression and diagnostics tests
R²	0.571
Adj-R²	0.557
F-statistics	8.321 (0.000)
DW statistics	1.930
Jarque-Bera normality test	2.992 (0.223)
ARCH hetero test	0.4702 (494)
B-G LM serial correlation test	0.182 (0.833)
CUSUM	Stable
CUSUMQ	Stable

Selected Model (2, 1, 1, 1, 1, 2, 1) based on AIC

Source: The author.

Note: ∆ is first difference operator. Figures in parenthesis contain p-values; *, ** and *** denotes statistical significance at 1%, 5% and 10%, respectively.

The positive coefficient of inflation at one period lag, which is also statistically significant reflects presence of inflation persistence.²¹ This might be due to inflationary expectations or actual price rigidities. The negative sign of inflation at lag two indicates that inflation series is mean reverting. The presence of inflation persistence further supports the adoption of inflation targeting framework under which inflation is considered as a nominal anchor.

The positive coefficient of fiscal deficit including at one period lag indicates that fiscal deficit growth has an inflationary impact in the short run. Mohanty and John (2015) and RBI (2012) strongly argue that fiscal deficit is one of the major causing factors behind inflation in India after the global financial crisis. In the short run it is not possible to increase output, so any fiscal stimulus might raises prices through excess demand pressures. It may be noted here that with limited scope for direct monetisation, the increase in fiscal deficit may generate inflationary pressures through rise in aggregate demand channel.²²

Diagnostic and Stability Test Results

The diagnostic tests are reported at the bottom of Table 6. The Jarque-Bera test statistics indicate that residuals of the model are normally distributed; there is no evidence of heteroscedasticity in the data as reflected by the ARCH test statistics. Further, the B-G LM serial correlation test statistics show there is no serial correlation in the residuals.²³ It may be seen from Figure 2 that the plots of the estimates of both CUSUM and CUSUMSQ tests are within critical bounds lines at the 5% level of significance. It means that our model is a good fit and all the estimates of parameters are stable over the sample period.

Figure 2.

Source: The author.

Granger Causality Analysis Using VECM

The causality test results are presented in Table 6. There exists a bidirectional causality between inflation and exchange rate in the long run, which means inflation and exchange rate are causing each other. Causality from the exchange rate to inflation indicates the presence of ERPT effect while causality from inflation to exchange rate may be due to the impact of domestic inflation on exports. Further, there is no evidence of money supply causing changes in the real output in the long run, which means money is neutral to output in the long run. This finding thus supports the proposition of Classical dichotomy, which suggests that in the long run, money is neutral to output.

Table 6.

Granger Causality Test Using VECM (Lags = 4)

	Direction of Causality
	Short Run (Wald-Test)							Long Run
Dependent variable/ regressors	$\sum_{i = 1}^{p} ⊿ I n π_{t - i}$	$\sum_{i = 1}^{p} ⊿ I n M 3_{t - i}$	$\sum_{i = 1}^{p} ⊿ I R_{t - i}$	$\sum_{i = 1}^{p} ⊿ I n Y_{t - i}$	$\sum_{i = 1}^{p} ⊿ I n F D_{t - i}$	$\sum_{i = 1}^{p} ⊿ I n E X R_{t - i}$	$\sum_{i = 1}^{p} ⊿ I n T O P_{t - i}$	EC_t–1 t-test
	Σθ_i=0 F-statistics	Σϕ_i=0 F-statistics	Σω_i=0 F-statistics	Σρ_i=0 F-statistics	Σχ_i=0 F-statistics	Σφ_i=0 F-statistics	Σv_i=0 F-statistics	ε=0 t-statistics
∆Inπ_t	–	3.306** (0.025)	2.756*** (0.0708)	4.175* (0.019)	2.378*** (0.077)	2.278*** (0.069)	6.602* (0.000	–0.321* [–2.880]
∆InM3_t	8.286* (0.000)	–	3.431** (0.022)	2.218 (0.118)	4.126** (0.021)	1.417 (0.2447)	0.562 (0.641)	–0.067 [–1.223]
∆IR_t	1.102 (0.354)	0.550 (0.650)	–	0.871 (0.460)	0.6929 (0.5595)	1.442 (0.245)	0504 (0.606	–0.455 [0–581]
∆InY_t	3.112*** (0.051)	8.058* (0.000)	3.361** (0.040)	–	2.450*** (0.0710	–0.036 [0.945	0.187 (0.829)	0.069 [1.445]
∆InFD_t	3.014* (0.003)	0.312 (0.847)	0.872 (0.460)	1.317 (0.276)	–	0.566 (0.572)	0.566 0.319	–3.422 [0.185]
∆InEXR_t	5.442** (0.042)	2.615*** (0.057)	2.252*** (0.089)	0.713 (0.495)	1.820 (0.172)	–	6.602* (0.000)	–0.318** [–2.393]
∆InTOP_t	2.561*** (0.081)	5.482* (0.006)	0.607 (0.612)	3.755** (0.029)	0.105 (0.899)	0.724 (0.765)	–	–0.076 [–1.342]

Source: The author.

Note: *, ** and *** denotes statistical significance at 1%, 5% and 10% level, respectively. The numbers in parenthesis are p-values and [] are t-values. ∆ is the first difference operator. The appropriate number of lags is based on AIC

Turning to the short run causality, there exists bidirectional causality between inflation and money supply. This implies that inflation and money supply are causing each other. This, bidirectional causality between money and prices means inflation leads to money creation thus rendering money targeting a complicated exercise. Several studies on the inflationary process in developing countries have argued that causality between money and prices may not be unidirectional, as postulated by the Monetarists model. This result seems reasonable because a rise in money supply causes a rise in prices via an increase in aggregate demand or purchasing power. This may also be possible that inflation leads to a rise in money supply. During periods of high inflation, the government’s expenditure is also likely to increase. Thus, it is possible that the government may resort to borrowing from RBI to meet its expenditure, which may increase the money supply. In this regard, the Aghevli–Khan hypothesis (Aghevli & Khan 1978) is especially important. According to this hypothesis, government expenditure adjusts more rapidly than receipts to a given change in price level and as a result, inflation widens the fiscal deficit through financing of the Central bank, leading to larger money supply and intensifying inflation further. Sarma (1982) reported that government expenditure get adjusted at a faster rate than government revenue to inflation, leading to higher budget deficits and, as a consequence, increasing money supply and price level.

Further, one-way causation exists from rate of interest to inflation, that is, changes in interest rate explains price level. This finding supports the Wicksell (1898) effect, which posits a negative association between interest rates and inflation rates with causality running from interest rates to inflation. The Wicksell effect plays an essential role in modern monetary policy formulation. This concept is used by the Central banks to control inflation by changing the interest rate due to the negative impact of interest rate on aggregate economic activity through investment and consumption channels and eventually, on inflation rates. This result further supports the adoption of CMR as an operating target in the current monetary policy action.

Also, the short run causality is running from interest rate and money supply to both inflation and output. This means that money is not neutral to output in the short run as believed by Monetarists. This finding implies that monetary policy in India has a role to play. The interest rate causes real output via changes in investment. The lower costs of capital cause investment spending and aggregate demand to rise, which may lead to higher economic growth. And it is with the help of interest rates that monetary authority brings changes in the money supply. This implies that interest rate is causing money supply and in turn money causes output.

Next, the fiscal deficit is found to granger cause real output in the short run. This seems logical because government expenditure in the case of developing economies like India is usually done on infrastructure projects, which stimulates investment and thus output. Hence, in case of India there is Crowding-in effect rather than Crowding-out effect. This result is also consistent with the neoclassical theory which suggests that fiscal stimulus stimulates economic growth.

Conclusion and Policy Implication

The primary purpose of the study is to investigate the causes of inflation in India over the period 1991–1992Q1 to 2017–2018Q4. Higher inflation has been always considered as one of the growth retarding factor and as a means of reducing the economic welfare of common masses. Therefore, maintaining price stability as defined by low inflation rate has remained a challenging task for the macroeconomic policy makers particularly for an emerging economy like India. Hence, identifying determinants of inflation is crucial for designing anti-inflationary policy measures. Under this background, the study has attempted to access the role of monetary variables, fiscal variables, supply side variables and external variables in explaining inflationary trends in India in the post reform period. The study has applied bounds testing procedure (developed within an ARDL model) to investigate the existence of long run equilibrium relationship between inflation and its determinants. Given that cointegration exists, a further two step procedure has been followed to estimate long run and short run elasticities under least squares method of an ARDL model. The direction of causal relationship has been tested by applying VECM based Granger causality test.

Historically, India has remained a moderate inflation country. The average annual WPI inflation rate for the 70 years period from 1951–1952 to 2020–2021 stood at 5.98%. The inflation record since planning has indicated that both inflation and inflation volatility declined gradually, particularly in the post reform period. This decline in inflation and its volatility suggested that monetary policy, to a large extent, has been effective in reducing inflation and inflation volatility.

The cointegrating relationship among variables has been established corresponding to inflation as dependent variable. The bounds test F-statistic value confirmed the existence of long run relationship (cointegration) among variables. The presence of cointegration allowed for estimation of the long run and short run elasticities. The long run and short run ARDL estimates has indicated that monetary variables such as money supply and nominal exchange rate are found to generate inflationary pressures while interest rate acts as an anti-inflationary policy instrument. The positive impact of money supply on inflation supports the Monetarists proportion that inflation is always a monetary process. The positive impact of exchange rate on inflations reflects the presence of ERPT. The output growth has negative impact on inflation and thus helps in mitigating inflation. The increasing fiscal deficit has inflationary impact only in the short run while in the long run the relationship between inflation and fiscal deficit is statistically insignificant. The positive relationship between inflation and fiscal deficit is in line with fiscal theory of price level. Finally, the positive relationship between inflation and trade openness refutes the validation of Romer (1993) and others in the Indian context who argue that openness acts as a mechanism to lower inflation.

These empirical findings suggests some policy recommendations. The finding that money and interest rate is causing output is in line with Monetarists and New-Keynesian economists who believe that money is not neutral to output at least in the short run. This also suggests that monetary policy in India has an important role to play in the growth process. The finding that that interest rate (CMR) mitigates inflation recommends that interest rate can be used as contractionary monetary policy instrument. Hence, the study supports the conduct of the current monetary policy framework in which CMR is recognized as an operating target. Given the close relationship between inflation and output growth, any policy which increases output is likely to have a favourable impact on inflation. Thus targeting supply-side constraints will help to moderate inflation. Adopting policies that improve productivity would also control inflation. Besides moderating prices, this will also reduce reliance on imports of those products for which domestic capacity exists. The findings also conclude that the money supply causes prices and output. So to ensure growth and price stability, the monetary authorities should focus on maintaining the adequate liquidity of rupee in line with the continuous credit flow to productive sectors of the economy. This implies that monetary authorities should also keep a close watch on monetary aggregates to ensure a stable price level. Finally, the positive link between inflation and openness has important policy implications in the Indian context, particularly for optimal trade policy. If the rising inflation discourages domestic capital formation and if capital accumulation is needed for development, it is therefore suggested that outward-looking trade policy (more openness) may not be optimal as it is inflationary. India being an importing economy; thus, more openness may make it more vulnerable to external shocks. Therefore, the optimal trade policy stance depends upon the stage of development.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Aghevli

B. B.

, & Khan

M. S.

(1978). Government deficits and the inflationary process in developing countries. Staff Papers, 25(3), 383–416.

Ahluwalia

M. S.

(2020). Backstage: The story behind India’s high growth years. Rupa Publications.

Ajaz

, Nain

M. Z.

, & Kamaiah

(2016). Inflation and openness in India: An asymmetric approach. Macroeconomics and Finance in Emerging Market Economies, 9(2), 190–203.

Ashra

, Chattopadhyay

, & Chaudhuri

(2004). Deficit, money and price: The Indian experience. Journal of Policy Modeling, 26(3), 289–299.

Balakrishnan

(1997). The fiscal deficit in macroeconomic perspective. Public Finance: Policy Issues for India.

Basu

(2011). Understanding inflation and controlling it. Economic & Political Weekly, 46(41), 50–64.

Bhat

J. A.

, & Sharma

N. K.

(2020). Identifying fiscal inflation in India: Some recent evidence from an asymmetric approach. Journal of Economics, Finance and Administrative Science, 25(50), 363–393.

Bhattacharya

, Patnaik

, & Shah

(2008). Exchange rate pass-through in India. National Institute of Public Finance and Policy.

Bhoi

B. K.

, Mitra

A. K.

, Singh

J. B.

, & Sivaramakrishnan

(2017). Effectiveness of alternative channels of monetary policy transmission: Some evidence for India. Macroeconomics and Finance in Emerging Market Economies, 10(1), 19–38.

10.

Brown

R. L.

, Durbin

, & Evans

J. M.

(1975). Techniques for testing the constancy of regression relationships over time. Journal of the Royal Statistical Society: Series B (Methodological), 37(2), 149–163.

11.

Cagan

(1956). The monetary dynamics of hyperinflation. In Friedman

(Ed.), Studies in the quantity theory of money (pp. 25–117). The University of Chicago Press.

12.

Callen

, & Chang

(1999). Modelling and forecasting inflation in India [Working Paper No 99/119]. International Monetary Fund.

13.

Chow

G. C.

(1960). Tests of equality between sets of coefficients in two linear regressions. Econometrica: Journal of the Econometric Society, 28(3), 591–605.

14.

Cooke

(2010). Openness and inflation. Journal of Money, Credit and Banking, 42(2–3), 267–287.

15.

Das

(2003). Modelling money, price and output in India: A vector autoregressive and moving, average (VARMA) approach. Applied Economics, 35(10), 1219–1225.

16.

Durbin

, & Watson

G. S.

(1950). Testing for serial correlation in least squares regression: I. Biometrika, 37(3/4), 409–428.

17.

Evans

R. W.

(2007). Is openness inflationary? Imperfect competition and monetary market power [Working Paper No 1]. Globalization and Monetary Policy Institute, Federal Reserve Bank of Dallas.

18.

Franses

P. H.

, & Franses

P. H. B.

(1998). Time series models for business and economic forecasting. Cambridge University Press.

19.

Jadhav

, & Singh

(1990). Fiscal–monetary dynamic nexus in India: An econometric model. Economic & Political Weekly, 25(3), 159–165.

20.

Jalil

, Tariq

, & Bibi

(2014). Fiscal deficit and inflation: New evidences from Pakistan using a bounds testing approach. Economic Modelling, 37, 120–126.

21.

John

R. M.

(2003). Inflation in India: An analysis using post-liberalisation data. https://dx.doi.org/10.2139/ssrn.470465.

22.

Joshi

A. R.

, & Acharya

(2010). Inflation and trade openness: Empirical investigation for India. IUP Journal of Monetary Economics, 8(1/2), 113–127.

23.

Kapur

, & Behera

H. K.

(2012). Monetary transmission mechanism in India: A quarterly model [Working Paper No 09]. Reserve Bank of India.

24.

Kaur

(2021). Inflation and fiscal deficit in India: An ARDL approach. Global Business Review, 22(6), 1553–1573.

25.

Khundrakpam

J. K.

(2008). Have economic reforms affected exchange rate pass-through to prices in India? Economic & Political Weekly, 43(16), 71–79.

26.

Khundrakpam

J. K.

, & Pattanaik

(2010). Fiscal stimulus and potential inflationary risks: An empirical assessment of fiscal deficit and inflation relationship in India. Journal of Economic Integration, 25(4), 703–721.

27.

Lane

P. R.

(1997). Inflation in open economies. Journal of International Economics, 42(3–4), 327–347.

28.

Leeper

E. M.

, & Yun

(2006). Monetary–fiscal policy interactions and the price level: Background and beyond. International Tax and Public Finance, 13(4), 373–409.

29.

Lin

H. Y.

, & Chu

H. P.

(2013). Are fiscal deficits inflationary? Journal of International Money and Finance, 32, 214–233.

30.

Maitra

(2016). Inflation dynamics in India: Relative role of structural and monetary factors. Journal of Quantitative Economics, 14(2), 237–255.

31.

Meltzer

A. H.

(2004). Origins of the great inflation. Federal Reserve Bank of St. Louis Review, 87(2, Part 2), 145–175.

32.

Mishra

, & Mishra

(2010). A VAR model of monetary policy and hypothetical case of inflation targeting in India. Monash University.

33.

Mohanty

, & Bhanumurthy

N. R.

(2014). Exchange rate regimes and inflation: Evidence from India. International Economic Journal, 28(2), 311–332.

34.

Mohanty

, & John

(2015). Determinants of inflation in India. Journal of Asian Economics, 36, 86–96.

35.

Nair

M. S.

(2014). Inflation dynamics in India: An analysis. Munich University.

36.

Narayan

P. K.

(2004). Fiji’s tourism demand: The ARDL approach to cointegration. Tourism Economics, 10(2), 193–206.

37.

Nguyen

(2015). Effects of fiscal deficit and money M2 supply on inflation: Evidence from selected economies of Asia. Journal of Economics, Finance and Administrative Science, 2(38), 49–53.

38.

Parida

P. C.

, Mallick

, & Mathiyazhagan

M. K.

(2001). Fiscal deficits, money supply and price level: Empirical evidence from India. Journal of Indian School of Political Economy, 13(4), 583–593.

39.

Patel

U. R.

, Chinoy

, Darbha

, Ghate

, Montiel

, Mohanty

, & Patra

(2014). Report of the expert committee to revise and strengthen the monetary policy framework. Reserve Bank of India.

40.

Pesaran

H. M.

, & Pesaran

(1997). Microflt 4.0. Oxford University Press.

41.

Pesaran

H. H.

, & Shin

(1998a). Generalized impulse response analysis in linear multivariate models. Economics Letters, 58(1), 17–29.

42.

Pesaran

M. H.

, & Shin

(1998b). An autoregressive distributed-lag modelling approach to cointegration analysis. Econometric Society Monographs, 31, 371–413.

43.

Pesaran

M. H.

, Shin

, & Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics, 16(3), 289–326.

44.

Radford

R. A.

(1945). The economic organisation of a POW camp. Economica, 12(48), 189–201.

45.

Ramu

M. A.

, & Gayithri

(2017). Fiscal deficit and inflation linkages in India: Tracking the transmission channels. Journal of Social and Economic Development, 19(1), 1–24.

46.

Reserve Bank of India (2012), Fiscal and monetary coordination. Report on currency and finance 2009–2012.

47.

Rogoff

(1992). Can international monetary policy cooperation be counterproductive? In International economic policies and their theoretical foundations (pp. 874–892). Academic Press.

48.

Romer

(1993). Openness and inflation: Theory and evidence. The Quarterly Journal of Economics, 108(4), 869–903.

49.

Saha

, & Zhang

(2013). Do exchange rates affect consumer prices? A comparative analysis for Australia, China and India. Mathematics and Computers in Simulation, 93, 128–138.

50.

Sarma

Y. S. R.

(1982). Government deficit, money supply and inflation in India. Reserve Bank of India Occasional Papers, 3(1), 56–67.

51.

Sabade

(2014). Is money supply the cause of inflation in India? An alternative postulate to understand inflation. Procedia—Social and Behavioral Sciences, 133, 379–382.

52.

Sharma

, & Mittal

(2019). Fiscal deficit, capital formation, and economic growth in India: a nonlinear ARDL model. Decision, 46(4), 353–363.

53.

Srinivasan

, Mahambare

, & Ramachandran

(2009). Dynamics of inflation in India: Does the new inflation bias hypothesis provide an explanation? Macroeconomics and Finance in Emerging Market Economies, 1(2), 199–212.

54.

US Census Bureau (2013). X-13ARIMA-SEATS reference manual [Technical Report].

55.

Wicksell

(1898). Interest and prices: A study of the causes of regulating the value of money. MacMillan.

56.

Woodford

(1994). Monetary policy and price level determinacy in a cash-in-advance economy. Economic Theory, 4(3), 345–380.

57.

Woodford

(1995). Price-level determinacy without control of a monetary aggregate. In Carnegie-Rochester conference series on public policy (Vol. 43, pp. 1)–46). North-Holland.

58.

Yadav

I. S.

, & Lagesh

M. A.

(2011). Macroeconomic relationship in India: ARDL evidence on cointegration and causality. Journal of Quantitative Economics, 9(1), 156–168.