Determinants of inflation in India
Click here to load reader
Transcript of Determinants of inflation in India
![Page 1: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/1.jpg)
Accepted Manuscript
Title: Determinants of inflation in India
Author: Deepak Mohanty Joice John
PII: S1049-0078(14)00067-0DOI: http://dx.doi.org/doi:10.1016/j.asieco.2014.08.002Reference: ASIECO 957
To appear in: ASIECO
Received date: 12-9-2013Revised date: 27-8-2014Accepted date: 30-8-2014
Please cite this article as: Mohanty, D., and John, J.,Determinants of inflation in India,Journal of Asian Economics (2014), http://dx.doi.org/10.1016/j.asieco.2014.08.002
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
![Page 2: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/2.jpg)
Page 1 of 29
Accep
ted
Man
uscr
ipt
Paper identifies the major determinants of inflation in India using a time varying
SVAR model for the period 1996-2014.
Influence of monetary policy on inflation remained almost steady during the study
period.
Output gap had an asymmetric impact on inflation with its influence having
weakened in the recent period.
Crude oil price was the predominant driver of inflation during 2009-11.
Fiscal deficit was a key determinant of inflation in 2011-12.
![Page 3: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/3.jpg)
Page 2 of 29
Accep
ted
Man
uscr
ipt
Title of the article: Determinants of inflation in India
Author names and affiliations: Deepak Mohanty Reserve Bank of India
Executive Director, 17th Floor, NCOB, Fort, Mumbai-400001, India, Tel.: +91-22-22633146, E-mail address: [email protected]
Joice John
Reserve Bank of India
Assistant Adviser, Department of Statistics and Information Management, C8/6th Floor, Bandra-Kurla Complex,
Mumbai-400051, India, Tel.: +91-22-26578315, E-mail address: [email protected]
Corresponding author: Joice John
Assistant Adviser, Reserve Bank of India, Department of Statistics and Information Management, C8/6th Floor, Bandra-Kurla
Complex, Mumbai-400051, India, Tel.: +91-22-26578315, E-mail address: [email protected]
![Page 4: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/4.jpg)
Page 3 of 29
Accep
ted
Man
uscr
ipt
2
1. Introduction
Historically, inflation in India had remained moderate. Average annual inflation rate
measured by the headline Wholesale Price Index (WPI) for the 62- year period from
1950-51 to 2012-13 was around of 6.7 percent. In recent years, prior to the global
financial crisis, from 2000-01 to 2007-08, average annual inflation was even lower at
around 5.2 percent. However, headline inflation rose close to 10 percent during 2010-
11 and 2011-12 before showing some decline in 2012-13. This sudden surge in inflation
and its persistence in the face of significant negative output gap was puzzling.
While stylized facts attribute the rise in inflation to both global and domestic factors, in
addition to supply and demand factors, there is hardly any systematic empirical study on
this aspect. This could perhaps be because of rapid changes in the drivers of inflation
over a short period of time. Against this backdrop, the paper attempts to identify the
determinants of inflation in India using a structural vector auto regression (SVAR)
model. Further, in order to capture the temporal changes in inflation dynamics within
the structural framework, a time varying parameter SVAR model with stochastic volatility
is estimated. The time-varying SVAR model has been extensively used in the recent
literature for examining the changes in macroeconomic dynamics and provides a
flexible approach where the parameters in equations as well as the volatility are
permitted to change over time [1,2].
The paper is organized as follows. A brief narrative on the determinants of inflation
describing the stylized facts in the Indian context is given in section 2. Econometric
framework is presented in section 3. Section 4 presents the empirical results. Section 5
concludes.
2. Determinants of inflation
Canova, Gambetti, and Pappa [3] using a time-varying SVAR found that there are many
similarities in the structural behaviour of inflation and output across a number of
advanced countries: in the US, changes in demand shocks were found to be most
significant; in the euro area, changes in the monetary policy shocks and supply shocks
were the major determinants; and in the UK, demand shocks, supply shocks and the
monetary policy were important. For the emerging market economies (EME), Mohanty
![Page 5: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/5.jpg)
Page 4 of 29
Accep
ted
Man
uscr
ipt
3
and Klau [4], using the data for the 1980s and 1990s found that supply factors, including
large changes in the external factors and agricultural shocks, drove domestic inflation,
while traditional demand factors even though significant were relatively weak. Unsal and
Osorio [5] studying inflation dynamics in Asia showed that the contribution of monetary
and supply shocks to inflation declined and domestic demand played a major role in
driving inflation in the 2000s.
In this paper, global commodity prices, output gap, fiscal policy and monetary policy are
examined as the key elements in the determination of inflation in India.
2.1 Global commodity prices
With the gradual external liberalisation, the Indian economy has more opened up than
ever before (Fig.1). Hence the global commodity prices and the exchange rate are
playing increasingly important role in the determination of inflation. Currently, near about
80 percent of crude oil demand in India is met by imports. During 2009 to 2011, global
commodity prices had an adverse impact on domestic inflation. In 2012, the
depreciation of the Indian Rupee more than offset the beneficial impact of marginal
decline in global commodity prices on domestic inflation (Fig.2) [6]. Further, some
studies on exchange rate pass through to domestic prices in the Indian context suggest
that 100 basis points (bps) change in the exchange rate had around 10 bps impact on
inflation [7,8,9].
2.2 Output gap
Traditional empirical work on the Phillips curve had tended to focus on the output gap
as a key indicator of inflationary pressures. The evidence on Phillips-curve relationship
in India is mixed. A number of studies covering data up to the 1990s or earlier did not
find the conventional Phillips-curve pattern [10,11,12,13,14,15]. However, Coe and
McDermott [16] found that the output gap is an important determinant of inflation in
almost all the Asian economies including India. Further, Paul [17] found that, in India,
after controlling for certain supply shocks, it is likely to have the short-run tradeoff
between inflation and industrial output.
![Page 6: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/6.jpg)
Page 5 of 29
Accep
ted
Man
uscr
ipt
4
2.3 Fiscal policy
Theoretical explanations of the fiscal impact of inflation are well postulated in the
macroeconomic theory. As argued by Sargent and Wallace [18], there is a dynamic
relationship between fiscal deficit and inflation. Since borrowing programs usually allow
governments to allocate ‘seigniorage’ over a period of time, the relationship between
fiscal deficit and inflation need not be contemporaneous. This might be a reason why
empirical research has had limited success in uncovering the relationship between fiscal
deficit and inflation [19,20]. However, Fischer, Sahay and Végh [21] using a panel data
of 94 countries, found that the fiscal deficits were one of the main drivers of high
inflation. Catao and Terrones [22] using a data set of 107 countries over the period of
1960–2001 found that the fiscal deficits were inflationary in most of the countries.
Further, this relationship was found to be especially strong for developing economies.
Lin and Chu [23] using a panel data of 91 countries for the period 1960 to 2006, found
that the fiscal deficit had a strong impact on inflation when inflation was high, and had a
weak impact when inflation was low. In a study on India covering the period 1953-2009,
Khundrakpam and Pattanaik [24] found that the fiscal deficit had a significant impact on
inflation. Traditionally the fiscal deficit in India remained relativity high. The fiscal
consolidation process during the period 2003-08 was reversed in 2008-09 mainly on
account of the financial crisis driven fiscal stimulus measures, which is postulated to
have had significant impact on inflation.
2.4 Monetary policy
Recent studies on monetary policy transmission in the Indian economy had found
increasing importance of the role of interest rates. It is also worth mentioning that from
May 2011 Reserve Bank of India has modified the operating procedure of the monetary
policy with a move to a single policy repo rate, with weighted average overnight call
money rate being explicitly recognized as the operating target of the monetary policy.
Anand, Peiris and Saxegaard [25] in a DSGE framework found that the peak impact of a
100 bps increase in the nominal policy rate was around 15 bps on inflation and it was
felt in the first quarter after the policy rate shock. Khundrakpam [26] observed that an
increase of 100 bps in the nominal policy rate was found to reduce bank credit by 2.2-
![Page 7: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/7.jpg)
Page 6 of 29
Accep
ted
Man
uscr
ipt
5
2.8 percent. Mohanty [27], using a quarterly SVAR model, found that the policy rate
increases had a negative effect on inflation; the peak effect on inflation was with a lag of
three quarters and the overall impact persisted through 8-10 quarters. Kapur and
Behera [9] found that 100 bps increase in the nominal policy rate lowered non-food
manufactured products’ inflation by 25 bps with a lag of 5 quarters.
3. Analytical framework
In order to assess the determinants of inflation a 5 variable SVAR is formulated, which
apart from inflation includes domestic crude oil prices, output gap, fiscal deficit and the
overnight weighted average call money rate. Inflation (πt) is measured as the
annualized growth rate of the de-seasonalised1 Wholesale Price Index (WPI). The WPI
based inflation was chosen for the study because during the study period (Q1: 1996-97
to Q3: 2013-14) the central bank was primarily focusing on WPI based inflation as the
main inflation measure, in absence of a nation-wide representative Consumer Price
Index (CPI). However, in this paper, for checking the robustness of the results, we also
used an alternate specification with inflation based on the annualized change in the de-
seasonalised1 Gross Domestic Product (GDP) deflator. Global commodity prices (gt) are
measured using the average crude oil price in Indian Rupees, which broadly represent
both international price and exchange rate pass-through to domestic inflation. This data
series is represented as the seasonally adjusted1 annualized growth rate. The output
gap (dt) is estimated by applying Hodrick-Prescott filter on de-seasonalised1 real Gross
Domestic Product (GDP). The ratio of seasonally adjusted1 gross fiscal deficit of the
central government to seasonally adjusted1 nominal GDP (at market prices) is used to
represent the fiscal policy (ft). The monetary policy (it) is represented using weighted
average call money rate, which is the operating target for monetary policy.
The structural identification restrictions for the SVAR estimation are as follows:
i) Crude oil price is considered to be the most exogenous to the framework.
ii) Inflation responds immediately to changes in crude oil price and to other
factors with a lag.
1Seasonal adjustment is done using X-12 ARIMA
![Page 8: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/8.jpg)
Page 7 of 29
Accep
ted
Man
uscr
ipt
6
iii) Output gap representing demand conditions is assumed to be sensitive to
international crude oil prices and domestic price movements.
iv) Fiscal deficit is sensitive to demand conditions and inflation.
v) Monetary policy is formulated considering the demand factors and price
situation besides being influenced by global factors and fiscal policy.
Let Yt denote a n ×1 vector gt, πt, dt, ft, it of 5 variables at time t. The structural
identifications described above can be incorporated in to a VAR framework by putting
the economic restrictions to draw inference about the structural relations. This can be
represented as:
AYt = F1Yt-1 +…+ FsYt-s + ut (1)
Where, ut is a (p=5) × 1 vector of structural shocks following N (0,Σ) where Σ = diag( σj).
For identifying this structural model, p(p + 1)/2 restrictions has to be imposed on A. Of
which p restrictions could be satisfied by normalizing the diagonal elements in A to
unity. The estimation in our case is with p=5, hence we have to impose 10 restrictions
on the contemporaneous correlations for identification of the five structural shocks. That
is straight away built into the system by the set of structural identification restrictions.
Thus A is given by a lower triangular matrix providing the 10 necessary restrictions for
identification.
Thus equation (1) could be rewritten as a reduced form VAR model.
Yt = B1Yt-1 +…+ BsYt-s + A-1Σεt (2)
Where εt follows N(0,I) and Bi = A-1Fi
Stacking the elements in the rows in Bi to form β and defining Xt = I ⊗(Yt-1'… Yt-s'), (2)
could be written as:
Yt = Xtβ + A-1Σεt (3)
The time varying dynamics of inflation determinants could be better understood if the
parameters of the model postulated in (3) are allowed to change over time. This is done
![Page 9: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/9.jpg)
Page 8 of 29
Accep
ted
Man
uscr
ipt
7
by analysing the structure using a Time-Varying Parameter Structural Vector Auto
Regression (TVP-VAR) framework, which is discussed in following section.
3.1 Time-varying parameter SVAR with stochastic volatility
In late 1990s TVP-VAR analysis became popular. Cogley and Sargent [28] used 3
variables with time-varying coefficients in a VAR model to study the persistence of
inflation in the US. Subsequently, Cogley and Sargent [29] introduced stochastic
volatility into the VAR model with time-varying coefficients but with a non-varying
structural shock. Primiceri [30] also used a three variable time varying VAR model,
which allowed all parameters to vary over time for the US economy. Cogley, Primiceri
and Sargent [1] used a time-varying VAR to identify persistence in inflation in the US
and defined a new measure of persistence. Benati and Surico [31] have also used
similar kind of approach to define the inflation persistence again for the US. Canova,
Gambetti and Pappa [3] used the time-varying SVAR to examine the dynamics of output
growth and inflation in the US, euro area and the UK. TVP-VAR was used for analyzing
time varying properties of the UK data by Benati [32]. Baumeister, Durinck and
Peersman [33] studied the effects of excess liquidity shocks on macroeconomic
variables in the euro area using a TVP-VAR model. TVP-VAR model for the Japanese
macroeconomic data was used by Nakajima, Kasuya and Watanabe [34] and Nakajima
[35]. Mumtaz and Plassmann [2] used TVP-VAR to study the time varying properties of
the real exchange rate for the UK, euro area and Canada.
All parameters in equation (3) are time-invariant. Following the methodology described
in Nakajima [36], this can be extended to TVP-VAR model by allowing the parameters
to change over time.
Yt = Xtβt + At-1Σtεt, t = s+1….n. (4)
Let at = (a21, a31, a32, a41, . . . , a54)' be a stacked vector of the lower-triangular elements
in At ; ht = (h1t, . . . , hkt)' with hjt = logσ2jt for j = 1, . . . , k, t = s+1, . . . , n and βt be the
vector of coefficients. Then following Primiceri [30] and Nakajima [36], it is assumed that
parameters in (4) follow a random walk process.
βt+1= βt +uβt
![Page 10: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/10.jpg)
Page 9 of 29
Accep
ted
Man
uscr
ipt
8
at+1= at +uat (5)
ht+1= ht +uht
with [εt, uβt, uat, uht]’ follows N(0, diag(I Σβ, Σa Σh)) for t = s+1, . .n.
Where , Σβ, Σa and Σh are the variance- covariance structure for the innovations of the
time-varying parameters. Following Cogley and Sargent [29] and Nakajima [36] Σβ, Σa,
and Σh are assumed to be diagonal. Further, following the related literature [1,30,36], a
tighter prior is set for Σβ and diffuse priors for Σa and Σh. The hyper parameters of Σβ are
simulated from an inverse Wishart distribution while Σa and Σh are drawn from an
inverse gamma distribution. Let Yt = ytt-1n; ω = (Σβ, Σa, Σh) and the prior density of ω be
π(ω). Given the data Y, the samples from the posterior distribution, π(β, a, h, ω|y) are
drawn following a Monte Carlo Marco Chain (MCMC) algorithm using the Matlab codes
developed by Nakajima [36]. The details of the estimation methodology are as available
in Nakajima [36].
4. Results
The 5 variable SVAR with the structural restrictions as imposed by the lower triangular
matrix A is first estimated in a time-invariant framework to understand the average
nature of dynamic relationship between the variables in the study period. In the
subsequent section the results of the TVP-VAR in the same framework are discussed.
Both estimations are done using the quarterly data from Q1:1996-97 to Q3:2013-14.
The choice of study period is on account of the availability of quarterly data on GDP and
to incorporate the post-liberalisation phase of the Indian economy, which assigned
greater role to market forces in determination of price, interest rate and exchange rate.
4.1 Empirical results from a time-invariant SVAR
The stationary properties of all the variables are tested and are found to be satisfactory
(Table 1). The lag length for the estimated SVAR is chosen to be 4 by the Schwarz
Bayesian Information Criterion (BIC). The regression diagnostics of no autocorrelation,
and homoscedasticity of residuals are found to be satisfactory. The impulse response
function (IRF) and forecast error variance decompositions (FEVD) are used to draw
conclusions on the determinants of inflation. The structural impulse response functions
![Page 11: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/11.jpg)
Page 10 of 29
Accep
ted
Man
uscr
ipt
9
with respect to one percentage point shock of different variables on inflation estimated
from the SVAR are presented in Fig.3. The 2-quarter, 4-quarter and 8-quarter ahead
structural FEVD are given in Table 2.
The impact of global factors represented by the domestic price of crude oil on inflation is
immediate and significant with the peak impact being in the first quarter. One
percentage point increase in the Indian basket crude oil price (in Rupees) leads to an
increase of 7 bps in domestic inflation accumulated over a period of 2 quarters. The
output gap is found to be insignificant at 95 per cent level of significance. The fiscal
deficit seems to impact WPI inflation with a longer lag of 5 quarters, with the peak
impact being estimated in the 5th quarter ahead. One percentage point increase in gross
fiscal deficit to GDP ratio is estimated to increase inflation by 106 bps over a period of 8
quarters. The peak impact of the policy rate on inflation is in the 4th quarter ahead. One
percentage point increase in the call money rate leads to 94 bps reduction in inflation
accumulated over a period of 8 quarters (Fig.3).
The Structural FEVD in Table 2 suggests that more than half the variation in inflation is
explained by its own shocks. Almost 22 percent of variation in WPI inflation 2 quarters
ahead is explained by the global factors: international crude oil price and exchange rate.
On an average around 3 percent of variation in inflation is explained by the output gap.
The fiscal deficit explains around 13 percent and the policy rate explains around 10
percent of variation in inflation in 8 quarters ahead.
4.2 Results with alternative measures of inflation
For checking the robustness of the above results we used an alternate specification with
inflation based on the GDP deflator. We did not use Consumer Price Index (CPI) based
measure of inflation because of the non-availability of a representative nation-wide CPI
series for India for the study period. The empirical results are more or less similar even
though the impact of various factors was found to be somewhat lesser in the deflator
based measure of inflation.
One percentage point increase in the Indian basket crude oil price (in Rupees) leads to
an increase of 4 bps in domestic inflation accumulated over a period of 2 quarters. The
output gap is not found to be significant at 95 per cent level of significance. The fiscal
![Page 12: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/12.jpg)
Page 11 of 29
Accep
ted
Man
uscr
ipt
10
deficit seems to significantly impact inflation in the 4th and 5th quarters. One percentage
point increase in gross fiscal deficit to GDP ratio is estimated to increase inflation by 76
bps over a period of 8 quarters. One percentage point increase in the call money rate
leads to 71 basis points reduction in inflation accumulated over a period of 8 quarters
with a significant impact in the 4th quarter (Fig.4).
As in the case with WPI inflation, the Structural FEVD suggests that over half the
variation in deflator-based inflation is explained by its own shocks. The output gap,
fiscal deficit and policy rate explain 4.8 per cent, 11.0 percent and 13.1 per cent of
variation respectively in the deflator based measure of inflation 8 quarters ahead (Table
3).
4.3 Empirical results from a time varying parameter SVAR with stochastic volatility
The TVP-VAR model with stochastic volatility described in equations (4) and (5) is used
to elucidate the time varying dynamics of various factors that explain inflation over the
study period. We followed the estimation procedure as available in Nakajima [36]. A
most commonly used method for setting the priors is laid down by Primiceri [30], where
priors are chosen based on the estimates of a time-invariant VAR model computed
using the pre-sample period. Even though it is reasonable to use this approach, due to
lack of sufficient data points in our sample we restricted the analysis by selecting a
reasonably flat prior for the initial state as we have not much information on the initial
state a priori [36]. The hyper parameters of Σβ, Σa and Σh are set as
Σβ~ Inverse Wishart (25, 0.01); (Σa)-2 ~ Gamma(4,0.02) and (Σh)
-2 ~ Gamma(4,0.02).
For the initial state of the parameters, flat priors are set with µβ0 = µa0 =µh0 = 0, and Σβ0
= Σa0 = Σh0 = 10 × I.
To compute2 the posterior estimates, 10,000 samples are drawn. However, the initial
1,000 samples are removed while calculating the posterior estimates. The sample paths
look stable and after the initial draws the sample autocorrelations are low. This indicates
that the sampling method produced the samples with low autocorrelation. The estimates
for the convergence diagnostics (CD) of Geweke [37], which are computed from the
2We used Matlab code developed by Nakajima [36] Available at (http://sites.google.com/site/jnakajimaweb/tvpvar.)
![Page 13: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/13.jpg)
Page 12 of 29
Accep
ted
Man
uscr
ipt
11
MCMC sample, shows that the convergence to the posterior distribution is not rejected
for the parameters. The inefficiency factors were found to be low indicating an efficient
sampling for the parameters.
Fig.5 provides the stochastic volatility of the five variables. Stochastic volatility of oil
prices showed a sharp increase in and around 2008-09 following the global financial
crisis. The volatility in Indian inflation exhibited a general downward trend after 2000.
The output gap showed some downward trend in volatility especially after 2003.
Volatility in call money rate remained at the same level, though with some temporal ups
and downs.
The time varying macroeconomic dynamics that determined Indian inflation during
1996-96 to 2012-13 are captured using the time varying IRFs and FEVDs of various
shocks on inflation in the TVP-VAR framework given in equations (4) and (5). The
accumulated impacts over 6 quarters ahead are presented in Fig. 6.
The impulse response functions reveal that the impact of global price shocks,
measured in terms of oil prices on domestic inflation was relatively high during 2009-10
and 2010-11. The accumulated impact of one percentage point positive shock in global
crude prices on inflation was around 12 bps during those periods. This came down to
about 5 bps in 2012-14. The output gap remained a notable determinant of inflation in
2007-08 and 2009-10. In 2007-08, the accumulated impact of one percentage point
positive shock to the output gap resulted in an increase of around 65 bps in inflation.
The impact got reduced in 2008-09, but went up to almost the earlier level in 2009-10.
Subsequently, this impact witnessed a general downward trend, with some cyclical ups
and downs. The fiscal shocks were relatively less adverse to inflation prior to 2005-06.
This, however, showed some upward movement in the six years from 2005-06 to 2012-
13 with some cyclical ups and downs. There was a considerable increase in the
accumulated impact of one percentage point shock to fiscal factors on inflation during
2011-12. The impact of the policy shocks on inflation remained almost unchanged
during the period 2005-06 to 2013-14 barring a few quarters in 2010-11. In the recent
period, one percentage point positive shock in call money rate accumulates to around
120 bps reduction in inflation in 6 quarters.
![Page 14: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/14.jpg)
Page 13 of 29
Accep
ted
Man
uscr
ipt
12
In order to examine which of the identified shocks are important to explain the inflation
fluctuations we decompose the forecast error variance of inflation into contributions from
the five identified shocks (Fig. 7). The time-varying nature of the SVAR model implies
that this decomposition can be calculated at each point in time.
Global oil price shock was the dominant driver of inflation during 2009-10 and most part
of 2010-11 with contribution above 20 percent. This declined to below 20 percent during
2011-14. During 2006-08, the output gap was contributing to inflation dynamics in the
range of 15 to 25 percent, which reduced during the financial crisis period. Its impact
again rose briefly in the first half of 2009-10 but declined to under 10 percent
subsequently. The fiscal factors were contributing under 10 percent almost throughout
the study period. However, in 2011-12 their contribution rose to about 10 percent, which
reduced subsequently during 2012-14. The monetary policy represented by call money
rate was found to contribute appreciably to inflation only after 2005-06. Since then, its
contribution was hovering in and around 10 to 20 percent barring a few quarters during
2009-11.
It is a challenge to develop a consistent empirical explanation of the recent inflation
dynamics in India, not surprisingly though as the economy is undergoing rapid structural
changes and experienced supply and demand shocks in the post-crisis period. We
estimated a time varying SVAR model for explaining the inflation in India, which better
captures the time varying properties of the inflation process through its proximate
determinants, than a standard SVAR model with the same determinants. From the
supply side, crude oil prices and exchange rate were found to be playing an important
role in defining domestic prices as the bulk of the country’s petroleum requirements are
met by imports. From the demand side, even though the output gap was found to be
insignificant on an average, it had an asymmetric impact on inflation: largely
contributing to rise in inflation when the output gap in high and positive but having only
trivial contribution to lower inflation when the output gap is negative. This could be
because in a supply constrained economy, even as firms operate below capacity they
still protect their pricing power to some extent. The fiscal deficit was found to have
contributed significantly to inflation during the post-crisis period. While both global and
domestic factors have played a role in the recent inflation process in India, the role of
![Page 15: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/15.jpg)
Page 14 of 29
Accep
ted
Man
uscr
ipt
13
monetary and fiscal policy becomes important in the containment of inflation given their
significant impact.
5. Conclusion
The paper tries to examine the factors that may have contributed to inflation in India in
the one and a half decades from 1996-97 to 2013-14 using an econometric framework
by incorporating the standard determinants of inflation viz.: i) crude oil prices, ii) output
gap, iii) fiscal policy iv) monetary policy and v) intrinsic inflation persistence. Further it is
also attempted to analyze the temporal changes in inflation dynamics in India within a
structural framework using a time varying parameter SVAR model. The paper
documents that inflation process in India in recent years has been complex with the
drivers changing frequently. The global price shocks, measured by domestic price of
crude oil that was predominant during 2009-11 have moderated in 2012-13. The output
gap has an asymmetric effect on inflation with its impact having weakened since 2012-
13, as the economy registered negative output gap. The fiscal deficit emerged as one
of the key determinants of inflation in 2011-12. The monetary policy impact on inflation,
however, has remained broadly unchanged. The paper underscores the role of
monetary and fiscal policy in the containment of inflation irrespective of the nature of the
shock to the inflation process.
Acknowledgement
The views expressed in the paper are those of the authors and do not represent those
of the institution to which authors belong. The authors would like to acknowledge the
useful comments received from two anonymous referees. However remaining errors
and omissions, if any, are those of the authors.
References
[1] T. Cogley, G.E. Primiceri, T.J. Sargent, Inflation-Gap Persistence in the US, Am.
Econ. J. Macroecon. 2(1) (2010) 43-69.
![Page 16: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/16.jpg)
Page 15 of 29
Accep
ted
Man
uscr
ipt
14
[2] H. Mumtaz, L.Sunder‐Plassmann, Time‐Varying Dynamics of the Real Exchange
Rate: An Empirical Analysis, J. Appl. Econ. (2012) doi: 10.1002/jae.2270.
[3] F. Canova, L. Gambetti, E. Pappa, The Structural Dynamics of Output Growth and
Inflation: Some International Evidence, The Econ. J. 117(519) (2007) C167-C191.
[4] M.S Mohanty, M. Klau, What determines inflation in emerging market economies?,
Bank of International Settlements Papers 8 (2001), Available
athttp://www.bis.org/publ/bppdf/bispap08a.pdf, Accessed on February 08, 2013.
[5] D.F Unsal, C. Osorio, Inflation Dynamics in Asia: Causes Changes and Spillovers
from China, J. Asian. Eco. 24 (2013) 26–40.
[6] Reserve Bank of India, Macroeconomic and Monetary Developments First Quarter Review 2013-14 (2013), Available athttp://rbidocs.rbi.org.in/rdocs/Publications/PDFs/M290713FC66587B599.pdf ,Accessed on July 30, 2013.
[7] R. Bhattacharya, I. Patnaik, A. Shah, Exchange rate pass-through in India, National
Institute of Public Finance and Policy: New Delhi(2008), Available at
http://macrofinance.nipfp.org.in/PDF/BPS2008_erpt.pdf, Accessed on February 08,
2013.
[8] J.K. Khundrakpam, Have Economic Reforms Affected Exchange Rate Pass-Through
to Prices in India?, Econ. Polit. Wkly. (2008) 71-79.
[9] M. Kapur, H. Behera, Monetary Transmission Mechanism in India: A Quarterly
Model, Reserve Bank of India Working Paper 09 (2012), Available at
http://rbidocs.rbi.org.in/rdocs/Publications/PDFs/WP09250612FL.pdf, Accessed on
February 08, 2013.
[10] S.S Bhalla, India's closed economy and world inflation, in: W.R. Cline and
Associates (Eds.), World Inflation and the Developing Countries, Brookings Institution,
Washington DC, 1981, pp.137.
[11] R.Chatterji, The behavior of industrial prices in India, Oxford University Press,
Delhi, 1989.
![Page 17: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/17.jpg)
Page 16 of 29
Accep
ted
Man
uscr
ipt
15
[12] S.K. Samanta, Price surprises and real output: The Indian evidence, Indian. Econ.
J. 34(2) (1986) 49–58.
[13] B.B. Bhattacharya, M. Lodh, Inflation in India: An analytical survey, ArthaVijnana
32(March) (1990) 25–68.
[14] R.H. Dholakia, Extended Phillips curve for the Indian economy, Indian.Econ.
J.38(1) (1990) 69-78.
[15] V. Virmani, Estimating Output Gap for the Indian Economy: Comparing Results
from Unobserved-Components Models and the Hodrick-Prescott Filter, Indian Institute
of Management Ahmadabad Research and Publication Department. No. WP2004-04-02
(2004), Available at http://www.iimahd.ernet.in/publications/data/2004-04-02vineet.pdf,
Accessed on February 08, 2013.
[16] D.T. Coe, C.J. McDermott, Does the gap model work in Asia?, Social Science
Research Network, Available at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=882967, Accessed on February 08,
2013.
[17] B.P. Paul, In search of the Phillips curve for India, J. Asian. Econ. 20(4) (2009) 479-
488.
[18] T.J. Sargent, N. Wallace, Some unpleasant monetarist arithmetic, Fed. Reserve.
Bank. Minneapolis. Q. Rev. 5(3) (1981) 1-17.
[19] R.G. King, C.I. Plosser, Money, deficits, and inflation, in: Carnegie-Rochester
Conference Series on Public Policy, Vol. 22, No. 1, Elsevier, North-Holland, 1985, pp.
147-195.
[20] R.W. Click, Seigniorage in a cross-section of countries, J. Money. Credit. Bank.
30(2) (1998) 154-171.
[21] S. Fischer, R. Sahay, C.A. Végh, Modern hyper-and high inflations, J. Econ. Lit. 40
(2002) 837-880.
[22] L. A. Catao, M.E. Terrones, Fiscal deficits and inflation, J. Monet. Econ. 52(3)
(2005) 529-554.
![Page 18: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/18.jpg)
Page 17 of 29
Accep
ted
Man
uscr
ipt
16
[23] H.Y. Lin, H.P. Chu, Are fiscal deficits inflationary?, J. Int. Money. Financ. 32 (2013)
214–233.
[24] J.K Khundrakpam, S. Pattanaik, Fiscal Stimulus and Potential Inflationary Risks: An
Empirical Assessment of Fiscal Deficit and Inflation Relationship in India, J. Econ.
Integr. 25(4) (2010) 703-721.
[25] R. Anand, S.J. Peiris, M. Saxegaard, An estimated model with macro-financial
linkages for India, International Monetary Fund Working Paper WP/10/21 (2010),
Available at http://www.imf.org/external/pubs/ft/wp/2010/wp1021.pdf, Accessed on
February 08, 2013.
[26] J.K. Khundrakpam, Credit Channel of Monetary Transmission in India–How
Effective and Long is the Lag?, Reserve Bank of India Working Paper 20 (2011),
Available at http://rbidocs.rbi.org.in/rdocs/Publications/PDFs/CCMTIR51211.pdf,
Accessed on February 08, 2013.
[27] D. Mohanty, Evidence on Interest Rate Channel of Monetary Policy Transmission in
India. Reserve Bank of India Working Paper 06 (2012), Available at
http://rbidocs.rbi.org.in/rdocs/Publications/PDFs/WPS6180512FL.pdf, Accessed on
February 08, 2013.
[28] T. Cogley, T.J. Sargent, Evolving post-world war II US inflation dynamics, in: B.S.
Bernanke and K. Rogoff(Eds.), National Bureau of Economic Research
Macroeconomics Annual 2001 Volume 16, MIT press, 2002, pp. 331-388.
[29] T. Cogley, T.J. Sargent, Drifts and volatilities: monetary policies and outcomes in
the post WWII US, Rev. Econ. Dyn. 8(2) (2005) 262-302.
[30] G.E. Primiceri, Time varying structural vector auto-regressions and monetary
policy, Rev. Econ. Stud. 72(3) (2005) 821-852.
[31] L. Benati, P.Surico, Evolving US monetary policy and the decline of inflation
predictability, J. Eur. Econ. Assoc. 6(2‐3) (2008) 634-646.
[32] L. Benati, The “great moderation” in the United Kingdom, J. Money.
Credit.Bank.40(1) (2008) 121-147.
![Page 19: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/19.jpg)
Page 18 of 29
Accep
ted
Man
uscr
ipt
17
[33] C. Baumeister, E. Durinck, G. Peersman, (2008).Liquidity, inflation and asset
prices in a time-varying framework for the euro area, Social Science Research Network,
Available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1290842, Accessed on
February 08, 2013.
[34] J. Nakajima, M. Kasuya, T. Watanabe, Bayesian analysis of time-varying
parameter vector autoregressive model for the Japanese economy and monetary policy,
J. Jpn. Int. Econ. 25(3) (2011) 225-245.
[35] J. Nakajima, S. Shiratsuka, Y. Teranishi, The effects of monetary policy
commitment: Evidence from time-varying parameter VAR analysis, Bank of Japan
Institute for Monetary and Economic Studies Working Paper 10-E-06 (2010), Available
at http://www.imes.boj.or.jp/english/publication/edps/2010/10-E-06.pdf, Accessed on
February 08, 2013.
[36] J. Nakajima, Time-varying parameter VAR model with stochastic volatility: An
overview of methodology and empirical applications, Bank of Japan Institute for
Monetary and Economic Studies Working Paper 11-E-09 (2011), Available at
http://www.imes.boj.or.jp/research/papers/english/me29-6.pdf, Accessed on February
08, 2013.
[37] J. Geweke, Evaluating the accuracy of sampling-based approaches to the
calculation of posterior moments, in: J. M. Bernardo, J. O. Berger, A. P. Dawid, and A.
F. M. Smith (Eds.), Bayesian Statistics Volume 4, Oxford University Press, New York,
1992, pp. 169–188.
![Page 20: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/20.jpg)
Page 19 of 29
Accep
ted
Man
uscr
ipt
18
Figure Captions
Fig. 1: Total trade to GDP ratio
Fig. 2: Annual percent change in crude prices and exchange rate (India Rupee per US
Dollar)
Fig. 3: Response of inflation (WPI) to one percentage point shock in different variables
Fig. 4: Response of inflation (GDP deflator) to one percentage point shock in different
variables
Fig. 5: Stochastic volatility in different variables
Fig.6: Time-varying accumulated impulse response function – Response of inflation to
one percentage point shock in other variables
Fig. 7: Time varying forecast error variance decomposition (over 6 quarters) of inflation
![Page 21: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/21.jpg)
Page 20 of 29
Accep
ted
Man
uscr
ipt
Based on authors’ calculations using data on ‘India’s foreign trade’ and ‘GDP at current market prices’ (Source: Database on Indian
Economy, Reserve Bank of India)
0
10
20
30
40
50
1970-7
1
1972-7
3
1974-7
5
1976-7
7
1978-7
9
1980-8
1
1982-8
3
1984-8
5
1986-8
7
1988-8
9
1990-9
1
1992-9
3
1994-9
5
1996-9
7
1998-9
9
2000-0
1
2002-0
3
2004-0
5
2006-0
7
2008-0
9
2010-1
1
2012-1
3
%
Total trade to GDP ratio
Fig. 1: Total trade to GDP ratio
![Page 22: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/22.jpg)
Page 21 of 29
Accep
ted
Man
uscr
ipt
Based on authors’ calculations using data on ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank of India)
and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund)
-80
-60
-40
-20
0
20
40
60
80
Mar-
09
Jun-0
9
Sep-0
9
Dec-0
9
Mar-
10
Jun-1
0
Sep-1
0
Dec-1
0
Mar-
11
Jun-1
1
Sep-1
1
Dec-1
1
Mar-
12
Jun-1
2
Sep-1
2
Dec-1
2
Mar-
13
Jun-1
3
Sep-1
3
Dec-1
3
%
Crude oil price Exchange rate (Rs./$)
Fig. 2: Annual percent change in crude prices and exchange rate
![Page 23: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/23.jpg)
Page 22 of 29
Accep
ted
Man
uscr
ipt
(a) Crude oil price shock
(b) Output gap shock
(c) Fiscal shock
(d) Monetary policy shock
Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95 % CI
-1.5
-1
-0.5
0
0.5
1
1.5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95 % CI
-1
-0.5
0
0.5
1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95 % CI
-1.5
-1
-0.5
0
0.5
1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95 % CI
Fig. 3: Response of inflation (WPI) to 1 percentage point shock
![Page 24: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/24.jpg)
Page 23 of 29
Accep
ted
Man
uscr
ipt
(a) Crude oil price shock
(b) Output gap shock
(c) Fiscal shock
(d) Monetary policy shock
Based on authors’ calculations using data on ‘GDP at current factor cost’ ,‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
-0.05
-0.03
-0.01
0.01
0.03
0.05
0.07
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95% CI
-1.5
-1
-0.5
0
0.5
1
1.5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95% CI
-0.8
-0.4
0
0.4
0.8
1.2
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95% CI
-1.5
-1
-0.5
0
0.5
1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
95% CI
Fig.4: Response of inflation (GDP) to 1 percentage point shock
![Page 25: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/25.jpg)
Page 24 of 29
Accep
ted
Man
uscr
ipt
Crude oil price
Inflation
Output gap
Fiscal deficit
Call money rate
Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central government’ ,
Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank of India) and ‘IMF
Primary Commodity Prices’ (Source: International Monetary Fund).
0.0
0.1
0.2
0.3
0.4
0.5
Q1:9
7-9
8
Q3:9
8-9
9
Q1:0
0-0
1
Q3:0
1-0
2
Q1:0
3-0
4
Q3:0
4-0
5
Q1:0
6-0
7
Q3:0
7-0
8
Q1:0
9-1
0
Q3:1
0-1
1
Q1:1
2-1
3
Q3:1
3-1
4 2.E-04
2.E-04
3.E-04
3.E-04
Q1:9
7-9
8
Q3:9
8-9
9
Q1:0
0-0
1
Q3:0
1-0
2
Q1:0
3-0
4
Q3:0
4-0
5
Q1:0
6-0
7
Q3:0
7-0
8
Q1:0
9-1
0
Q3:1
0-1
1
Q1:1
2-1
3
Q3:1
3-1
4 4.E-06
4.E-06
5.E-06
5.E-06
Q1:9
7-9
8
Q3:9
8-9
9
Q1:0
0-0
1
Q3:0
1-0
2
Q1:0
3-0
4
Q3:0
4-0
5
Q1:0
6-0
7
Q3:0
7-0
8
Q1:0
9-1
0
Q3:1
0-1
1
Q1:1
2-1
3
Q3:1
3-1
4
9.0E-06
9.5E-06
1.0E-05
Q1:9
7-9
8
Q3:9
8-9
9
Q1:0
0-0
1
Q3:0
1-0
2
Q1:0
3-0
4
Q3:0
4-0
5
Q1:0
6-0
7
Q3:0
7-0
8
Q1:0
9-1
0
Q3:1
0-1
1
Q1:1
2-1
3
Q3:1
3-1
4
1.9E-07
2.0E-07
2.0E-07
2.1E-07
2.1E-07
Q1:9
7-9
8
Q3:9
8-9
9
Q1:0
0-0
1
Q3:0
1-0
2
Q1:0
3-0
4
Q3:0
4-0
5
Q1:0
6-0
7
Q3:0
7-0
8
Q1:0
9-1
0
Q3:1
0-1
1
Q1:1
2-1
3
Q3:1
3-1
4
Fig. 5: Stochastic volatility in different variables
![Page 26: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/26.jpg)
Page 25 of 29
Accep
ted
Man
uscr
ipt
(a) Crude oil price shock
(b) Output gap shock
(c) Fiscal shock
(d) Monetary policy shock
Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
-0.10
0.00
0.10
0.20
0.30
Q1:0
6-0
7
Q4:0
6-0
7
Q3:0
7-0
8
Q2:0
8-0
9
Q1:0
9-1
0
Q4:0
9-1
0
Q3:1
0-1
1
Q2:1
1-1
2
Q1:1
2-1
3
Q4:1
2-1
3
Q3:1
3-1
4
%
0.0
0.2
0.4
0.6
0.8
1.0
Q1:0
6-0
7
Q4:0
6-0
7
Q3:0
7-0
8
Q2:0
8-0
9
Q1:0
9-1
0
Q4:0
9-1
0
Q3:1
0-1
1
Q2:1
1-1
2
Q1:1
2-1
3
Q4:1
2-1
3
Q3:1
3-1
4
%
0.0
0.2
0.4
0.6
0.8
1.0
Q1:0
6-0
7
Q4:0
6-0
7
Q3:0
7-0
8
Q2:0
8-0
9
Q1:0
9-1
0
Q4:0
9-1
0
Q3:1
0-1
1
Q2:1
1-1
2
Q1:1
2-1
3
Q4:1
2-1
3
Q3:1
3-1
4
%
-1.6
-1.2
-0.8
-0.4
0.0
Q1:0
6-0
7
Q4:0
6-0
7
Q3:0
7-0
8
Q2:0
8-0
9
Q1:0
9-1
0
Q4:0
9-1
0
Q3:1
0-1
1
Q2:1
1-1
2
Q1:1
2-1
3
Q4:1
2-1
3
Q3:1
3-1
4
%
Fig.6: Time-varying accumulated impulse response function
![Page 27: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/27.jpg)
Page 26 of 29
Accep
ted
Man
uscr
ipt
Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
0.0
20.0
40.0
60.0
80.0
100.0
Q1:0
6-0
7
Q2:0
6-0
7
Q3:0
6-0
7
Q4:0
6-0
7
Q1:0
7-0
8
Q2:0
7-0
8
Q3:0
7-0
8
Q4:0
7-0
8
Q1:0
8-0
9
Q2:0
8-0
9
Q3:0
8-0
9
Q4:0
8-0
9
Q1:0
9-1
0
Q2:0
9-1
0
Q3:0
9-1
0
Q4:0
9-1
0
Q1:1
0-1
1
Q2:1
0-1
1
Q3:1
0-1
1
Q4:1
0-1
1
Q1:1
1-1
2
Q2:1
1-1
2
Q3:1
1-1
2
Q4:1
1-1
2
Q1:1
2-1
3
Q2:1
2-1
3
Q3:1
2-1
3
Q4:1
2-1
3
Q1:1
3-1
4
Q2:1
3-1
4
Q3:1
3-1
4
%
Inflation Crude oil price Output gap Fiscal deficit Call money rate
Fig. 7: Time varying forecast error variance decomposition
![Page 28: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/28.jpg)
Page 27 of 29
Accep
ted
Man
uscr
ipt
Table 1: Phillips-Perron test for unit root
Variables Test statistic p-value*
Inflation^ -WPI -5.468 0.000 Inflation^ - GDP Deflector -5.872 0.000 Crude oil price change** -5.849 0.000 Output gap -2.724 0.0699 Fiscal deficit -6.806 0.000 Policy rate -4.112 0.001
*MacKinnon approximate p-value
^ Measured as annualized quarterly percent change in seasonally adjusted data.
** Measured as annualized quarterly percent change in seasonally adjusted domestic prices.
For all variables expect output gap, unit root rejected at 1% level of significance. In case of output gap unit root is
rejected at 7% i.e. Stationary at 7% level of significance .
Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at current factor cost’, ‘GDP at constant
factor cost’, ‘Fiscal deficit of central government’ , Weighted average call money rate’ and ‘Average forex rates’
(Source: Database on Indian Economy, Reserve Bank of India) and ‘IMF Primary Commodity Prices’ (Source:
International Monetary Fund).
Table 1: Phillips-Perron test for unit root
![Page 29: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/29.jpg)
Page 28 of 29
Accep
ted
Man
uscr
ipt
Table 2: Forecast error variance decomposition – WPI inflation
Variables Variation in inflation explained- 2 quarters ahead (%)
Variation in inflation explained- 4 quarters ahead (%)
Variation in inflation explained- 8 quarters ahead (%)
Inflation (WPI)
72.1 63.7 55.8
Crude oil price 22.0 20.1 19.2 Output gap 2.0 2.1 3.1 Fiscal deficit 1.7 4.8 12.6 Policy rate 2.2 9.3 9.3 Based on authors’ calculations using data on ‘Wholesale Price Index’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
Table 2: Forecast error variance decomposition - WPI inflation
![Page 30: Determinants of inflation in India](https://reader038.fdocuments.in/reader038/viewer/2022100504/5750a1101a28abcf0c90b029/html5/thumbnails/30.jpg)
Page 29 of 29
Accep
ted
Man
uscr
ipt
Table 3: Forecast error variance decomposition – GDP deflator inflation
Variables Variation in inflation explained- 2 quarters ahead (%)
Variation in inflation explained- 4 quarters ahead (%)
Variation in inflation explained- 8 quarters ahead (%)
Inflation (GDP deflator)
73.6 54.8 51.3
Crude oil price 21.5 21.2 19.8 Output gap 3.4 3.2 4.8 Fiscal deficit 0.5 7.7 11.0 Policy rate 1.0 13.1 13.1 Based on authors’ calculations using data on ‘GDP at current factor cost’, ‘GDP at constant factor cost’, ‘Fiscal deficit of central
government’ , Weighted average call money rate’ and ‘Average forex rates’ (Source: Database on Indian Economy, Reserve Bank
of India) and ‘IMF Primary Commodity Prices’ (Source: International Monetary Fund).
Table 3: Forecast error variance decomposition - GDP inflation