Dissertation Report on

“Study the impact of Monetary Policy on

Commodity Price”

Submitted in Partial Fulfilment of the Requirement of the

Dissertation in the Master of Business Administration Programme
(Full Time) 2009-11

Author: Ashish Sarkhel (091309)

Faculty Guide: Prof Kshamanidhi Adabar

Date of Submission: Jan 11th 2011

Institute Name: Institute of Management, Nirma

University ,Ahmedabad
Table of Contents
Table of Contents........................................................................................................2
List of Figures............................................................................................................. 5
List of Tables............................................................................................................... 6
Abbreviation............................................................................................................... 6
Letter of Approval ......................................................................................................7
Acknowledgment........................................................................................................ 8
EXECUTIVE SUMMARY.................................................................................................9
Chapter 1.................................................................................................................. 10
Introduction.......................................................................................................... 10
Background........................................................................................................... 12
Literature Review.................................................................................................. 13
Objectives:............................................................................................................. 18
Chapter 2.................................................................................................................. 19
Research Plan....................................................................................................... 19
Selection of Monetary Indicator............................................................................19
Selection of Statistical Method..............................................................................22
Methodology......................................................................................................... 22
Hypothesis - Filtering random parameter ............................................................22
Develop the Model for Individual parameter .......................................................23
Develop the Model with all parameters.................................................................23
Data Source (Secondary Data).............................................................................23
Monetary Indicator:............................................................................................... 23
Commodity Indices................................................................................................24
NCDEX (National Commodity and Derivative Exchange Ltd.)................................24
1.1.1.1.MCX (Multi Commodity Exchange)............................................................24
Tools..................................................................................................................... 25
Pre Regression Data Analysis...............................................................................25

2| Page
 Post Regression Analysis......................................................................................25
Hypothesis:............................................................................................................ 25
Establishment of the Model...................................................................................26
Assumption........................................................................................................... 26
Methodological Issues...........................................................................................26
Limitation of Model...............................................................................................27
Data Analysis:........................................................................................................27
Identification of Trend Data..................................................................................27
Exchange Rate Analysis.........................................................................................27
Consumer Price Index Analysis..............................................................................28
Money Supply Analysis..........................................................................................29
Index of industrial Production Analysis..................................................................30
Interest rate Analysis.............................................................................................31
Regression Analysis..............................................................................................32
MCX....................................................................................................................... 33
Regression of MCXCOMDEX along with Exchange rate ........................................34
Regression of MCXCOMDEX along with CPI ..........................................................35
Regression of MCXCOMDEX along with Money Supply .........................................36
Regression of MCXCOMDEX along with IIP ............................................................37
Regression of MCXCOMDEX along with IIP ............................................................38
Multiple Regression of Commodity Price with all Monetary Factors.......................39
MCX Sub indices – Metal, Energy, Agri...................................................................40
Correlations: MCXCODMEX, SUB INDICES..............................................................40
Multiple Regression of MCX Agri with all Monetary Factors...................................41
NCDEX................................................................................................................... 42
Correlation between Dhanya (NCDEX) and MCX Agri ...........................................42
Chapter 3.................................................................................................................. 43
Interpretation........................................................................................................ 43
Exchange Rate......................................................................................................43
Consumer Price Index (CPI)...................................................................................43
Money Supply (M)..................................................................................................43
Production Index (IIP)............................................................................................44
Interest Rate.......................................................................................................... 44

3| Page
 Recommendation.................................................................................................. 44
Conclusion............................................................................................................. 45
Bibliography.......................................................................................................... 47
Appendix............................................................................................................... 50
Analysis of NCDX (“Ganya”)..................................................................................50
TRENDS IN INDEX OF INDUSTRIAL PRODUCTION...................................................53
Monthly Average Data for All Monetary Indicators................................................53

4| Page
List of Figures

Figure 2-1: Movement of Exchange Rate Along with Time........................................27

Figure 2-2 Autocorrelation Graph for Exchange Rate .............................................28
Figure 2-3 Movement of Consumer Price Index Along with Time..............................28
Figure 2-4 Autocorrelation Graph for Consumer Price Index....................................29
Figure 2-5 Movement of Money Supply Along with Time.........................................29
Figure 2-6 Autocorrelation Graph for Money Supply.................................................30
Figure 2-7 Movement of Index of Industrial Production Along with Time.................30
Figure 2-8 Autocorrelation Graph for IIP..................................................................31
Figure 2-9 Movement of Prime Lending Rate (Interest rate) Along with Time.........31
Figure 2-10 Autocorrelation Graph for Prime Lending Rate.....................................32
Figure 2-11 Movement of MCX Index (MCXCODMEX) Along with Time.....................33
Figure 2-12 Residual for Regression between MCXCODMEX and Exchange Rate.....34
Figure 2-13 Residual for Regression between MCXCODMEX and Consumer Price
Index.........................................................................................................................35
Figure 2-14 Residual for Regression between MCXCODMEX and Money Supply.......36
Figure 2-15 Residual for Regression between MCXCODMEX and Index of industrial
Production.................................................................................................................37
Figure 2-16 Residual for Regression between MCXCODMEX and Prime Lending Rate
.................................................................................................................................38
Figure 2-17 Residual for regression of MCXCODMEX with CPI, Money Supply, IIP,
Interest rate, Ex Rate................................................................................................39
Figure 2-18 Movement of Sub-indices of MCX along with Time................................40
Figure 2-19 Residual for regression of MCX Agri with CPI, Money Supply, IIP, Interest
rate, Ex Rate.............................................................................................................41
Figure 2-20 Movement of Dhanya (NCDEX) index along with time...........................42

5| Page
List of Tables

Table 2-1 Calculation of Consumer Price Index.........................................................20

Table 2-2 Weightage various Sector in Index of Industrial Production.....................20
Table 2-3 A sample of exchange rate arious currency against Indian Rupee..........21
Table 2-4 Component of various MCX sun indices and Main Index MCXCODMEX.....25

Abbreviation

RBI – Reserve Bank of India

IMF – International Monetary Fund

CPI – Consumer Price Index

IIP – Index of Industrial Production

Interest Rate – Prime Lending Rate

M – Money Supply (M1 Level)

MCX – Multi Commodity Exchange Ltd

NCDEX – National Commodity Exchange Ltd.

6| Page
Letter of Approval

Institute of Management, Nirma University

MBA (Full-Time)–II (2009-11 Batch)

Dissertation Report Approval Form

Roll No. : 091309

Student’s Name : Ashish Sarkhel

Faculty Guide : Prof Kshamanidhi Adabar

Topic : Study the impact of Monetary Policy on Commodity Price

Signature of Student

Approval of the Faculty Guide

7| Page
Acknowledgment

I would like to thank my research guide Prof Kshamanidhi Adabar, who has guided me through my
project despite his busy schedule. He was always there to listen and to give advice. He is responsible
for involving me in this interesting project in the first place. He showed me different ways to
approach a research problem and the need to be persistent to achieve results.

I would also like to thank to Prof P.K. Chugan and Prof Rajesh Jain who provided insightful
comment and feedback during proposal presentation, as these comment comes very useful to keep
the research resourceful.

I would also like to thank other faculty member and my peer group, who always support and
guide me in my dissertation report.

Place: Ahmedabad Ashish Sarkhel

Date: 11th Jan 2011

8| Page
EXECUTIVE SUMMARY

Commodities play an important role in the economies of most of the developing Countries,
which derive the majority of its merchandise export revenues from one single commodity or
several commodities. Against the background of rising commodity prices, most Asian economies
developed favourably between 1999 and 2005, although differences between net oil-exporting
and importing countries were apparent. Net agri commodity exporters recorded the highest
growth rates, mainly supported by rising investment and exports on the back of record
production and an expanding demand in some countries. Monetary policy helping to fight
inflationary.

In India exchanged based commodity price system has been newly developed and
exchange like MCX and NCDEX now making an important role in commodity market system.
Now the policy maker needs to consider the Commodity price also as due to various Monetary
Indicator Commodity Price change significantly. In this dissertation report two phases are there.
First, an analysis based on a system of five variables (commodity price index, money supply,
interest rates, consumer price index and industrial production index) to Indian economical Data,
emulating the study performed earlier by various authors using data from the RBI. Our second
step in this paper is to develop the complete model with all-variable system by multiple
regressions, in order to check the robustness of the empirical results obtained under the five-
variable system. Most of the empirical literature devoted to the assessment of the relationship
between monetary policy and commodity prices has focused on the interest rate as an indicator of
monetary policy stance. However, interest rates may not fully represent the impact of a monetary
policy shock and, more importantly, their movements can repeat the endogenous response of
monetary policy to the general developments of the economy.

So there is a need of a Mathematical Model which can show the significance of Monetary
Indicators in Commodity Pricing.

9| Page
 Chapter 1
Introduction

Many studies have debated whether the commodity price index serves as a leading indicator of
future economic conditions. Many authors have applied time series analysis to investigate how
this index relates to macroeconomic variables. Specifically, they applied statistical tests to
analyse the validity of the commodity price index as a leading indicator of prices in general. If
the commodity price index is valid as a leading indicator of prices or other macroeconomic
variables, policymakers need to recognize the importance of the index as a policymaking
variable. A rise in the index, for example, could be seen as a portent of future inflation. As such,
trends in the index would come to have a certain influence in the management of monetary
policy. Commodity prices are important as macroeconomic discussions were dominated by the
oil price shocks and other rises in agricultural and mineral products that were thought to play a
big role in the stagflation of that decade. Any discussion of alternative monetary regimes was not
complete without a consideration of the gold standard and proposals for other commodity-based
standards. Yet the topic of commodity prices fell out of favour in the late 1980s and the 1990s.
Commodity prices generally declined during that period; perhaps declining commodity prices are
not considered as interesting as rising prices. Nobody seemed to notice how many of the victims
of emerging market crises in the 1990s were oil producers that were suffering, among other
things, from low oil prices or others suffering from low agricultural prices.
The crisis that began with turmoil in the US subprime mortgage market in 2008 has
morphed into a crisis of two halves. In the first, rising inflation was a major concern, the
economic outlook, while weaker, was expected to remain robust, especially for emerging East
Asia. In the second-half, the opposite has happened. Amid falling world commodity prices
driven by a severely weakened global economy, policymakers have aggressively eased policies
to forestall further economic and financial deteriorations. This report focuses on the first-half of
the crisis. In the first-half, central banks in the region faced a major dilemma in rising consumer
prices and slower output growth. Higher interest rates would restrain rising inflation by lowering
domestic demand, but also hurt output growth. Many central banks were uncertain how to
respond. The challenge was rather a new experience: for a long time, monetary policy played a
supportive role to economic growth with inflation never too menacing given rising productivity,

10 | P a g e
plenty of spare capacity, and a generally disciplined fiscal disposition. The effect is not only
specific to few countries, India also need to look as commodity market a major indicator for
Monetary Policy. The establishment of commodity exchange like MCX and NCDEX shows
commodity shows a lot of indicator for future monetary action. The report tries to establish a
relationship between major Monetary Policy tools along with Commodity Indices.
It would be foolish to think that the model captures everything. In reality, a lot of other
things beyond real interest rates influence commodity prices. There are bound to be fluctuations
both in the long-run equilibrium real price, the convenience yield, storage costs and risk
premium. These fluctuations are not readily measurable. Such factors as weather, political
vicissitudes in producing countries, and so forth, are likely to be very important when looking at
individual commodities. Indeed analysts of oil or coffee or copper pay rather little attention to
macroeconomic influences, and instead spend their time looking at microeconomic determinants.
Oil prices were high in 2004-06 in large part due to booming demand from China and feared
supply disruptions in the Middle East, Russia, Nigeria and Venezuela. There may now also be a
premium built in to the convenience yield arising from the possibilities of supply disruption to
terrorism, uncertainty in the Persian Gulf, and related risks. Yet another factor concerns the
proposition that the world supply of oil may be peaking in this decade as new discoveries lag
behind consumption, that such predictions have in the past been proven wrong. This would imply
that the long-run equilibrium real price of oil has shifted upward. Other factors apply to other
commodities. In coffee, the large-scale entry of Vietnam into the market lowered prices sharply a
few years ago. Corn, sugar, and cotton are heavily influenced by protectionist measures and
subsidies in many countries. And so on. If we look at aggregate indices of commodity prices,
many of the idiosyncratic factors in individual markets wash out.
Commodity price developments have been one of the major sources of concern for
policymakers during the recent years. After having surged with increasing momentum to
unprecedented levels in the course of 2008, prices of commodities fell abruptly in the wake of
the financial crisis and the global economic downturn. Since the beginning of 2009, however,
commodity prices first stabilised, and then resumed an upward path, characterized by relatively
high volatility. As commodity prices in general .and the oil price in particular .are an important
component of Consumer Price Indexes, the evolution of these prices and the driving forces
behind them are clearly crucial for the conduct of monetary policy

11 | P a g e
Background
Commodity price developments have been one of the major sources of concern for
policymakers during the recent years. A wide strand of literature has examined the impact of
commodity prices on macroeconomic variables and the stance of monetary policy. Fewer
attention has however been devoted to the other direction of causality, i.e. the impact of
monetary conditions on oil and other commodity prices.
During the commodity price surge of 2008 some commentators indicated that loose
monetary policy and persistently low interest rates could have at least in part fuelled the price
hike. If this is so, it is then relevant to understand whether and to what extent the massive
monetary policy easing which is nowadays taking place may sow the seeds for another surge in
commodity prices. The aim of this study is indeed to analyze to what extent an expansionary
monetary policy change may drive up commodity prices and through which channel.
Monetary conditions and interest rates have attracted attention as possible driving factors
of commodity prices. Most of the empirical literature devoted to the assessment of the
relationship between monetary policy and commodity prices has focused on the US interest rate
as an indicator of monetary policy stance. However, interest rates may not fully represent the
impact of a monetary policy shock and, more importantly, their movements can reflect the
endogenous response of monetary policy to the general developments of the economy.
Our strategy is to identify a monetary policy change for other factor in the Indian
economy, and then assess its impact on commodity prices. This allows us not only to examine
the impact of monetary policy net of other interaction channels, but also to avoid employing
indicators of global monetary conditions which are inherently difficult to measure. The main
finding is that there is empirical evidence of a significant impact of monetary policy on
commodity prices; in particular, an expansionary monetary policy change drives up the broad
commodity price index and all of its major components. Taking a cue from Japanese Economy,
there the causal relationship from the BOJ index to the consumer price index changed sharply
when the Bank of Japan introduced its zero interest rate policy in February 1999. The BOJ index
was valid as a leading indicator of the consumer price index before the policy was introduced,
whereas the relationship ceased to exist thereafter. We found no causal relationship from the BOJ
index to the industrial production index. Note that this relationship is unrelated to BOJ’s
introduction of the zero interest rate policy. We can therefore argue that the commodity price

12 | P a g e
index has functioned less effectively as an information variable for monetary policy since the
adoption of the zero interest rate policy.
Commodity prices and the general price level tend to be closely related, with movements in
commodity prices leading movements in the general price level. Commodity prices are
determined in auction markets, hence they reflect demand and supply shocks more rapidly than
do the prices of manufactured goods. The change in commodity prices resulting from speculative
purchases or sales of commodities can therefore be a leading indicator of change in the general
price level. The price mechanism of financial markets was presumably no longer working when
the zero interest rate policy was introduced. This would make it impossible to consider the future
movements of monetary policy, and thus the CPI, when commodity prices are determined in
auction markets. Secondly, the Japanese economy was in a serious depression when the zero
interest rate policy was introduced
The variance decomposition suggests however that the impact of monetary policy on
commodity prices is rather limited, though statistically significant. Still, attempt is to throw
some light on the channel through which monetary policy change affect commodity prices,
focusing on the all major commodity. In particular, study investigate whether the positive impact
on commodity prices of a monetary policy loosening can be ascribed to incentives to stock
accumulation, disincentives to immediate production or to financial results.

Literature Review

Since the late 1990s, commodity prices have followed an upward trend, with the prices of
Metals and crude oil showing the most pronounced increases. Although booms in commodity
prices could be observed previously, the magnitude of the increase, its duration and its breadth
are unparalleled compared with other upswings in the past 25 years. Notably, prices for all
commodity groups have risen simultaneously since 2002, a pattern which cannot be detected for
such a prolonged period any time since 1980. Now oil prices and many broader indices
of commodity prices are again at or near all-time highs in nominal terms, and
are very high in real terms as well. Copper, platinum, nickel and zinc, for
example, all hit record highs in 2006, in addition to crude oil. As a result,
commodities are once again hot. It turns out that mankind has to live in the

13 | P a g e
physical world after all! Still, the initial reaction in 2003-04 was relaxed, on
several grounds:

1. Oil was no longer a large share of the economy, it was said;

2. Futures markets showed that the “spike” in prices was expected to be
only temporary; and
3. Monetary policy need focus only on the core CPI inflation rate and can
safely ignore the volatile food and energy component, unless or until it
starts to get passed through into the core rate.
With regard to point (3), it is time to examine more carefully the claim that if an
increase in energy or agricultural prices does not appear in the core CPI, then
monetary policy can ignore it. The central argument of our research, reported in
the NBER working paper “The Effect of Monetary Policy on Real Commodity
Prices,” is that high real commodity prices can be a signal that monetary policy
is loose. Thus they can be a useful monetary indicator (among many others).
The analysis is both theoretical and empirical. The empirical work includes the
determination of real commodity prices in the United States, the determination
of prices in other smaller countries, and the determination of inventories. But by
2005-2006, the increase in prices had gone far enough to receive much more
serious attention. This was especially true with regard to the perceived
permanence of oil prices, largely because the futures price had gone from
implying that the rise in the spot price was mostly temporary to implying that it
is mostly permanent.Instability of commodity prices has always been a major
concern of the producers, processors, merchandisers as well as the consumers
in an agriculture-dominated country like India. Farmers ‘direct exposure to price
fluctuations, for instance, makes it too risky for them to invest in otherwise
profitable activities. There are various ways to cope with this problem. Apart
from increasing the stability of the market by direct government intervention,
various actors in the farm sector can better manage their activities in an
environment of unstable prices through derivatives markets mainly futures and
options on futures. These markets serve a risk-shifting function, and can be used
to lock-in prices in advance instead of relying on uncertain price developments

14 | P a g e
in future. Apart from being a vehicle for risk transfer among hedgers and from
hedgers to speculators, these markets also play a major role in price discovery.
Svensson in his paper(2005) mention that commodity price developments have been one
of the major sources of concern for policymakers during the recent years. After having surged
with increasing momentum to unprecedented levels in the course of 2008, prices of commodities
fell abruptly in the wake of the financial crisis and the global economic downturn. Since the
beginning of 2009, however, commodity prices first stabilized, and then resumed an upward
path, characterized by relatively high volatility. As commodity prices in general and the oil price
in particular are an important component of Consumer Price Indexes, the evolution of these
prices and the driving forces behind them are clearly crucial for the conduct of monetary policy
A wide strand of literature has examined the impact of commodity prices oil in particular
on macroeconomic variables (e.g. Kilian, 2008), but fewer attention has been devoted to the
other direction of causality, i.e. the impact of monetary conditions on oil and other commodity
prices. In this study focus on the latter, to analyse to what extent an expansionary monetary
policy change may drive up commodity prices and through which channel.
Hamilton (2009) explained, while supply and demand factors can in general explain the
bulk of the fluctuations in commodity prices, other forces may at times play a role. Kilian (2009)
and Alquist and Kilian (2010) highlight the relevance of precautionary demand shocks, which
increase current demand for oil due to an increase in uncertainty about future oil supply
shortfalls, in the behaviour of oil prices. Since the seminal contribution by Frankel (1984),
monetary conditions and interest rates have attracted attention as possible driving factors of
commodity prices. Frankel (1986) extends statistical theory of exchange rate overshooting to the
case of commodities and, using no-arbitrage conditions, derives a theoretical link between oil
prices and interest rates.
Barsky and Kilian (2002, 2004) show that monetary policy stance is a good predictor of
commodity prices. In particular, Barsky and Kilian (2002) also suggest that the oil price
increases of the 1970s could have been caused, at least in part, by monetary conditions. Most of
the empirical literature devoted to the assessment of the relationship between monetary policy
and commodity prices has focused on the US interest rate as an indicator of monetary policy
stance (Frankel, 2007, Frankel and Rose, 2009). However, interest rates may not fully represent
the impact of a monetary policy shock and, more importantly, their movements can react the

15 | P a g e
endogenous response of monetary policy to the general developments of the economy. For
instance, Bernanke, Gertler and Watson (1997), using a regression framework, suggest that
positive shocks to the oil price induce a monetary policy response which can amplify the
contractionary effects of the oil price shock itself; Kilian and Lewis (2009), however, report no
evidence of systematic Fed reaction to oil shocks after 1987.
During the commodity price surge of 2008 some commentators indicated that loose
monetary policy and persistently low interest rates could have at least in part fuelled the price
hike (Hamilton, 2009). If this is so, it is then relevant to understand whether and to what extent
the massive monetary policy easing which is nowadays taking place may sow the seeds for
another surge in commodity prices. In this paper, study will not work with a plain analysis of co
movements between commodity prices and interest rates, but rather identify a monetary policy
change in a regression system for the Indian economy, and then assess its impact on commodity
prices. This allows us not only to examine the impact of monetary policy net of other interaction
channels, but also to avoid employing indicators of global monetary conditions which are
inherently difficult to measure. More specifically, study will use a standard identification scheme
for the monetary policy shock (Kim, 1999) and study will then project each of the commodity
prices on this shock in order to single out the responses of the different prices to the same
monetary policy shock. Study finds empirical evidence of a significant impact of monetary
policy on commodity prices; in particular, an expansionary monetary policy shock drives up the
broad commodity price index and all of its major components.
Although the methodology is very different, our approach is similar in spirit to that of
Frankel and Hardouvelis (1985), which investigated the impact of money supply announcements
on commodity prices; the main methodological difference lies in the fact that in our case study
work with an identified monetary policy shock in a regression system. In addition, study assesses
the robustness of the results by repeating the exercise using several different identification
strategies of the monetary policy shock, which are commonly used in the literature. In particular,
remaining in a regression context, study used also the Choleski identification strategy proposed
by Boivin and Giannoni (2006) and that based on sign restrictions in the spirit of Uhlig (2005)
and Canova and De Nicolò (2002). The actual implementation of the sign restrictions is obtained
through the algorithm developed in Rubio-Ramírez et al. (2010). Study also analysed the effect

16 | P a g e
on commodity prices of the monetary policy shocks identi.ed according to Kuttner (2001) and
Romer and Romer (2004). Overall, all these different strategies lead to similar conclusions.
The variance decomposition suggests that the impact of monetary policy on commodity
prices is rather limited, though statistically significant. Still, study try to shed some light on the
channel through which monetary policy shocks affect commodity prices, focusing on the case of
oil. In particular, study investigate whether the positive impact on oil prices of a monetary policy
loosening can be ascribed to incentives to stock accumulation, disincentives to immediate
production or to financial results. Results show that all these direct channels display the expected
sign, but the bulk of the impact of monetary policy on commodity prices seems to transit through
the indirect channel of expected growth and inflation, as also reported by Barsky and Kilian
(2004).
The impact of monetary policy on commodity prices has been studied by Barsky and
Kilian (2002, 2004), who argue that the channel through which monetary policy exerts its impact
on commodity prices is via (expectations of) stronger inflation and economic growth. There are
however a number of other channels, related to the opportunity cost of investing in real assets,
according to which an expansionary monetary policy can cause an increase in commodity prices.
Frankel (2007) summarizes them as:
1. Low interest rates tend to reduce the opportunity cost of carrying inventories,
increasing their demand for commodities;
2. On the supply side, lower rates create an incentive not to extract today
exhaustible commodities, as the cost of holding inventories in the ground also
decreases;
3. For a given expected price path, a decrease in interest rates reduces the carrying
Cost of speculative positions, making it easier to bet on assets such as
commodities; under certain conditions, this will put upward pressure on futures
price and, by arbitrage, also on spot prices.
The plan of the paper is the following. In section 2 paper evaluate the impact of monetary policy
shocks on the commodity price index and on its major components. In particular, study first
describes the theoretical arguments according to which commodity prices should react to
monetary policy changes. Next, study present the data and the econometric framework, which
includes the identification scheme, and then study provide regression evidence and robustness

17 | P a g e
analysis. Paper also evaluates the transmission channels through which monetary policy may
affect commodity prices.

Objectives:

1. To Identify relationship between various Monetary Indicators and Commodity Prices

1. Verify that the Indicator have a trend and useful for analysis
2. Establish the Model between the Indicator and Commodity Price
3. Interpret the result for All sub Commodity indices with Monetary Indicators
4. Establish a model between all indicators and Commodity Price

2. Forecast future action for Monetary Policy on the basis of Commodity Prices
1. Interpret the Effect of adverse condition of Monetary policy on Commodity Price
2. Generate preventive measure for fluctuation

The objective of paper is to develop a model which shows the relationship between change in
Monetary Policy and effect on Commodity Prices. In actual terms exact relationship is difficult
to establish and Commodity price is based on various parameters including Monetary Policy, but
in this report the trend has been analysed. It gives an insight how important is the commodity
sectors is for Indian Economy and Policy formulation. Application of Granger causality tests to
analyses the validity of the commodity price index as a leading indicator of prices in general. If
the commodity price index is valid as a leading indicator of prices or other macroeconomic
variables, policymakers need to recognize the importance of the index as a policymaking
variable. A rise in the index, for example, could be seen as a portent of future inflation. As such,

18 | P a g e
trends in the index would come to have a certain influence in the management of monetary
policy.

Chapter 2
Research Plan

Selection of Monetary Indicator

In order to develop the relationship between monetary policy Indicator and Commodity Index
Following Data has been collected for Monetary Indicator
1. Consumer Price Index (CPI)
Consumer Price Index (CPI) in India comprises multiple series classified based on
different economic groups. There are four series, viz the CPI UNME (Urban Non-Manual
Employee), CPI AL (Agricultural Labourer), CPI RL (Rural Labourer) and CPI IW
(Industrial Worker). While the CPI UNME series is published by the Central Statistical
Organisation, the others are published by the Department of Labour
The most common formula used in calculating consumer price indexes is a weighted
arithmetic mean of price relatives. The price relatives described earlier are weighted
according to the amounts consumers spend on each product; the resulting figures are
summed for all commodities and divided by the sum of the base year expenditures for the
same collection of commodities.

Major Group/Group Weights

Primary Food Article 32.295

Article Non-food Article 17.386
Mineral 10.081

19 | P a g e
 4.828
Fuel, Power, Light & Lubricant 10.663

Food Product 57.042

Beverages,
Tobacco & 10.143
Manufacture Tobacco
d Product Product 2.149
Textile 11.545
Leather & Leather
1.01
Product
Other 32.18

TOTAL 100
Table 2-1 Calculation of Consumer Price Index

2. Industrial Index of Production (IIP)

In India, the Central Statistical Organisation (CSO) is responsible for compilation and
release of the Index of Industrial Production (IIP). This is a monthly index and is
intended to measure changes over time in the volume of industrial production. The base
year of the current series of IIP in India is 1993-94 which is being revised to 1999-2000.
The current series of IIP with base 1993-94 is based on 538 individual items clubbed into
283 groups of items. The distribution of these items (item groups) and weights (100)
among the three sectors covered by the index is as under:

No. of Items (item Weigh

Sector
groups) t
Mining 64 (1) 10.47
Manufacturi
ng 473 (281) 79.36
Electricity 1 (1) 10.17
Total 538 (283) 100
Table 2-2 Weightage various Sector in Index of Industrial Production

3. Interest Rate
Here interest rate has been consider as Prime lending rate define by RBI
The interest rate that commercial banks charge their best, most credit-worthy customers.
20 | P a g e
 Generally a bank's best customers consist of large corporations. The rate is determined by
the Federal Reserve's decision to raise or lower prevailing interest rates for short-term
borrowing. Though some banks charge their best customers more and some less than the
official prime rate, the rate tends to become standard across the banking industry when a
major bank moves its prime up or down. The rate is a key interest rate, since loans to
less-creditworthy customers are often tied to the prime rate. For example, a Blue Chip
company may borrow at a prime rate of 5%, but a less-well-established small business
may borrow from the same bank at prime plus 2, or 7%. Many consumer loans, such as
home equity, automobile, mortgage, and credit card loans, are tied to the prime rate.
Although the major bank prime rate is the definitive "best rate" reference point, many
banks, particularly those in outlying regions, have a two-tier system, whereby smaller
companies of top credit standing may borrow at an even lower rate.
4. Exchange Rate
Here we have considered Rs/USD exchange rate.
Few example of exchange rate are as follows
Currency 1 INR in INR
American Dollar 0.0220353 45.3818
Argentine Peso 0.090559 11.0425
Australian Dollar 0.0221714 45.1031
Brazilian Real 0.0372953 26.813
British Pound 0.0141407 70.7179
Canadian Dollar 0.02187 45.7247
Chilean Peso 10.99 0.0909922
Chinese Yuan 0.146288 6.83584
Colombian Peso 41.1863 0.0242799
Croatian Kuna 0.126052 7.93324
Danish Krone 0.126783 7.88752
Euro 0.0170234 58.7426
Hong Kong Dollar 0.171388 5.8347
Table 2-3 A sample of exchange rate arious
currency against Indian Rupee

5. Money Supply
The four measures of money supply for annual compilation developed in India by the
SWG (1977) are as follows:
M1 = currency with public + demand deposits with the banking system + other deposits with RBI
M2 = M1 + saving deposits with post office savings banks
M3 = M1 + time deposits with the banking system

21 | P a g e
 M4 = M1 + all deposits with post office savings banks excluding National Saving Certificates
Here we consider only M1 as money supply as that part is major money supplier entity
among all. Considering all entity will be too tedious task.
The following data has been collected as these are the major factor and quantifiable in nature. So
a mathematical relationship is feasible.

Selection of Statistical Method

Various Researchers applied the Multiple Vector regression model to analyze the relationship
between the commodity price index and macroeconomic variables in the United States. They
found a causal relationship from the commodity price index to the consumer price index and the
industrial production index. This study empirically analyses the role of commodity prices in
India as a leading indicator of macroeconomic variables, especially the CPI (consumer price
index), post the period of year 2000. For study Vector Regression model has been used to
empirically analyse the relationship between the Commodity Price Index and macroeconomic
variables.

Methodology
In order to established a model for relationship appropriate need to be recognize, so all available
data for Monetary Indicator need to be auto regress ,so that only those indicator will be chosen
those are not random in nature

Hypothesis - Filtering random parameter

Establish Auto Correlation of all the Monetary Indicator and Verify following hypothesis
Monetary Indicator is not based on its past value(Random in nature)

H0: r =0
v/s
Ha: r # 0
Where r is Auto Correlation Coefficient
If the Hypothesis turns fail, only in that case the parameter will be used for relationship Model

22 | P a g e
 Develop the Model for Individual parameter
Once Indicators are filtered out, relationship establishment can be done initially each individual
parameter need to be regress with various commodity indices Here Regression implies, both data
will generate following Eqn

Yt = e + K1CPI(t) where t = 1, 2,3…..T

Yt : Forecasted Value of Commodity Index

e: Error Term(intercept)
K1: Slope of Eqn (Coefficient for CPI to be related with Commodity Index)
CPI (t): Value of CPI at t

Various Monetary Indicator need to be regress along all available indices to analysis individual
sector effect. Time Series Regression can be accomplished by either Unit Root Test or (Vector)
Regression. Here Vector Auto regression will be used, where Time Series data will be consider
as a vector

Develop the Model with all parameters

Finally a complete eqn established will all monetary indicators for predicating Commodity Index
value
Yt = e + K1CPI(t) + K2 IR(t) + K3IIP(t) + k4EX(t) + k5M(t) where t = 1,2,3…..T
Yt : Forecasted Value of Commodity Index
e : Error Term(intercept)
K1: Slope of Eq (Coffcient for CPI to be related with Commodity Index)
CPI(t) : Value of CPI at t
IR(t) : Value of Interest rate at time t
IIP(t) : Value of Index of Industrial Production at time t
EX(t) : Value of Exchange rate at time t
M(t) : Money Supply at time t

Data Source (Secondary Data)

Monetary Indicator:

1. Consumer Price Index (CPI) – www.rbi.org /Data warehouse

CPI data are directly available in Monthly basis

23 | P a g e
 2. Industrial Production Index - www.rbi.org /Data warehouse
IIP data are directly available on monthly basis
3. Interest Rate- www.rbi.org /Data warehouse
Here Prime Lending rate has been Consider as Interest rate and the data is based on Daily
basis, so monthly average has been taken for the given period. The interest rate is not
changing with a cycle manner, so even monthly data doesn’t show a particular trend.
Quarterly data can be useful, but as all other Monetary Indicators are in Monthly format
so Prime lending rate(Interest Rate) is also taken as Monthly one.
4. Exchange Rate - www.rbi.org /Data warehouse
Here Rs. /USD exchange rate has been consider as Exchange rate.These data is available
on daily basis ,so it need to average out for monthly basis.

Commodity Indices
For Commodity Price various Indices have been chosen from Commodity Exchange.

NCDEX (National Commodity and Derivative Exchange Ltd.)

Indices Name: Dhanya
Weightage:
Components Weight (%)
Chana 0.1135
Cotton Seed Oilcake 0.0203
Guar Seed 0.1711
Gur 0.0337
Jeera 0.0406
Pepper 0.052
Mustard Seed 0.1156
Soy bean 0.1522
Turmeric 0.0275
Wheat 0.2735

1.1.1.1. MCX (Multi Commodity Exchange)

Indices Name: MCX COMDEX
Sub-Indices: MCX METAL, MCX ENERGY,MCX AGRI
Weightage:
MCX COMDEX 2009 Weights
Weight
MCX COMDEX Commodity Group Adjusted Wts.
(New)

MCX METAL INDEX 15.21 40.00%

Gold
%
Silver 9.66%

24 | P a g e
 Copper 7.13%
Zinc 2.00%
Aluminum 2.00%
Nickel 2.00%
Lead 2.00%
35.41
Crude Oil
MCX ENERGY INDEX % 40.00%
Natural Gas 4.59%
Ref. Soy Oil 3.91%
Potato 4.76%
MCX AGRI INDEX Chana 4.14% 20.00%
Crude Palm Oil 3.19%
Kapaskhalli 2.00%
Mentha Oil 2.00%

Table 2-4 Component of various MCX sun indices and Main Index MCXCODMEX

Here we have taken commodity spot price and it is closing day value for all the Indices.

Tools
In order to gather the data in Structure Form MS-Excel 2010 has been used. All Statistics
Analysis e.g. Auto Correlation, Regression, Unit Root Test has been done in Minitab Ver. 15

Pre Regression Data Analysis

Data has been Analysis in following steps
1. Use Auto Correlation Analysis to see that all Monetary Indicators have a trend and they
are not random in nature. If any variable do not satisfy auto regression criteria then it will
not be used
2. Before Implementing the regression all-time series data are adjusted for common time
period and time space. For e.g. Commodity Index value is Weekly and CPI is monthly,
then an average for Commodity Index on Monthly basis need to be calculated.

Post Regression Analysis

Quantitative: On the basis of Regression Coefficient we can determine that whether there
is a relationship between monetary indicator and Commodity Price

Hypothesis:
We need to test two hypotheses here
1. “Monetary Indicator is not Related with Commodity Price”

25 | P a g e
 H0 : R2 = 0
v/s
Ha : R2 # 0
2. If the test fail to accept the above hypothesis then check
“Monetary Indicator is loosely associate with Commodity Price”
H0 : R2 low (< 50%)
v/s
Ha : R2 High (>50 %)

Establishment of the Model

The theoretical model is based on exchange rate overshooting and can be summarized as
follows. A monetary contraction temporarily raises the real interest rate, whether via a
rise in the nominal interest rate, a fall in expected inflation, or both. Real commodity
prices fall. In the long run, the general price level adjusts to the change in the money
supply. As a result, the real money supply, real interest rate, and real commodity price
eventually return to where they were.

In this study we will use the hypothesis to justify that overshooting model is applicable
for how many Monetary Indicators

Assumption

• Aggregate demand is determined by the standard open economy IS-LM mechanism

• Financial Markets are able to adjust to shocks instantaneously, and investors are risk
neutral.
• In the short run, goods prices are 'sticky'. That is, aggregate supply is horizontal in the short
run, though it is positively sloped in the long run.

Methodological Issues

The key issue in during the research is the appropriate statistical Model and relevant data.
Various tools for time series analysis demand and appropriate de seasonal and de trend data for
regression, while in order to achieve true effect of Monetary Indicator, data should be used

26 | P a g e
without any adjustment as Commodity price can be effected by factors which affect Monetary
Indicators e.g. political, Climatic etc.

Limitation of Model

• The Model can only explain the partial effect on Commodity price as Commodity Price is
also majorly derive by demand –supply of particular product
• For ease of calculation Vector Analysis has been simplified for same time period and all
data are average out for this , so very precise calculation may generate some errors
• In this model error lagging part has been removed for easy interpretation, which indicate
that if residual are increasing in nature then model can produce wrong results, unlike
ARIMA which consider the error part also.

Data Analysis:

Identification of Trend Data

Exchange Rate Analysis

Figure 2-1: Movement of Exchange Rate Along with Time

27 | P a g e
 Auto Correlation for Exchange Rate

1.0
0.8
0.6

0.4
Autocorrelation

0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

2 4 6 8 10 12 14 16 18 20 22 24 26 28
Lag

Figure 2-2 Autocorrelation Graph for Exchange Rate

As we can see that Autocorrelation Coefficient r # 0, so we can reject the

hypothesis and it implies that Exchange rate data have a trend and it depend on its
previous value ,So we can use it as regression for Commodity Index.

Consumer Price Index Analysis

Figure 2-3 Movement of Consumer Price Index Along with Time

28 | P a g e
 Auto Correlation for CPI

1.0
0.8
0.6
0.4
Autocorrelation

0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

1 5 10 15 20 25 30 35 40 45
Lag

Figure 2-4 Autocorrelation Graph for Consumer Price Index

As we can see that Autocorrelation Coefficient r # 0, so we can reject the

hypothesis and it implies that Consumer Price Index data have a trend and it
depend on its previous value, so we can use it as regression for Commodity Index.

Money Supply Analysis

Figure 2-5 Movement of Money Supply Along with Time

29 | P a g e
 Auto Correlation for Mone Supply

1.0
0.8
0.6
0.4
Autocorrelation

0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

2 4 6 8 10 12 14 16 18
Lag

Figure 2-6 Autocorrelation Graph for Money Supply

As we can see that Autocorrelation Coefficient r # 0, so we can reject the

hypothesis and it implies that Money Supply data have a trend and it depend on its
previous value, so we can use it as regression for Commodity Index.

Index of industrial Production Analysis

Figure 2-7 Movement of Index of Industrial Production Along with Time

30 | P a g e
 Auto Correlation for IIP

1.0
0.8
0.6
0.4
Autocorrelation

0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

1 5 10 15 20 25 30 35 40 45 50
Lag

Figure 2-8 Autocorrelation Graph for IIP

As we can see that Autocorrelation Coefficient r # 0, so we can reject the

hypothesis and it implies that Money Supply data have a trend and it depend on its
previous value, so we can use it as regression for Commodity Index.

Interest rate Analysis

Figure 2-9 Movement of Prime Lending Rate (Interest rate) Along with Time

31 | P a g e
 Auto Correlation for Interest Rate(Prime Lending Rate)

1.0
0.8
0.6
0.4
Autocorrelation

0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0

1 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Lag

Figure 2-10 Autocorrelation Graph for Prime Lending Rate

As we can see that Autocorrelation Coefficient r # 0, so we can reject the

hypothesis and it implies that Prime Lending rate data have a trend and it depend
on its previous value, so we can use it as regression for Commodity Index.
After analysis the pre regression data we can conclude that all five parameter
are fit for regression with Commodity Exchange Indices Data

Regression Analysis

In order to see the effect on commodity market we choose two commodity Indices
• MCX
• NCDEX

32 | P a g e
 MCX

Figure 2-11 Movement of MCX Index (MCXCODMEX) Along with Time

The MCX has following Indices

• MCXCOMDEX
• MCXENERGY
• MCXMETAL
• MCXAGRI

The initial regression start with MCXCOMDEX, in this index data comes daily, so it has been
average out for Month basis. Now all Monetary Indicator Need to regress against this Index
values on Monthly Basis

33 | P a g e
 Regression of MCXCOMDEX along with Exchange rate

Versus Fits
(response is MCX)
1000

500
Residual

-500

-1000
2150 2200 2250 2300 2350 2400 2450 2500 2550
Fitted Value

Figure 2-12 Residual for Regression between MCXCODMEX and Exchange Rate

Result

The regression equation is

MCX = - 701 + 66.1 EX
R2 = 7.7%
Hypothesis test

• Testing first Hypothesis

H0 : R2 = 0
As R2 = 7.7 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%
As R2 = 7.7 % < 50 % so the test fail to reject the Hypothesises Exchange rate is loosely
associated with Commodity Price
Now from the Equation it appears that value of Commodity Price is positive related with
Exchange rate (EX),but as R2 value is quite low (<50 %) ,so the relationship is not very strong.

34 | P a g e
 Regression of MCXCOMDEX along with CPI

Versus Fits
(response is MCX)
1000

500
Residual

-500

-1000
2000 2200 2400 2600 2800 3000
Fitted Value

Figure 2-13 Residual for Regression between MCXCODMEX and Consumer Price Index

Result

The regression equation is

MCX = 552 + 4.07 CPI
R2 = 44.0%
Hypothesis test

• Testing first hypothesis

H0 : R2 = 0
As R2 = 44 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%
As R2 = 44 % < 50 % so the test fail to reject the Hypothesises CPI is loosely associated
with Commodity Price
Now from the Equation it appears that value of Commodity Price is positive related with
Consumer Price Index (CPI),but as R2 value is quite low (<50 %) ,so the relationship is
not very strong, but it has a larger impact than exchange rate.

35 | P a g e
 Regression of MCXCOMDEX along with Money Supply

Versus Fits
(response is MCX)
1000

500
Residual

-500

2000 2200 2400 2600 2800 3000

Fitted Value

Figure 2-14 Residual for Regression between MCXCODMEX and Money Supply

Result

The regression equation is

MCX = 1171 + 0.00200 Money Supply
R2 = 48.8%
Hypothesis test

• Testing first hypothesis

H0 : R2 = 0
As R2 = 48.8 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%
As R2 = 48.8 % < 50 % so the test fail to reject the Hypothesises, Money Supply is
loosely associated with Commodity Price
Now from the Equation it appears that value of Commodity Price is positive related with
Money Supply, but as R2 value is just low (<50 %) ,so the relationship is not very strong,
but it has a larger impact than exchange rate.

36 | P a g e
 Regression of MCXCOMDEX along with IIP

Versus Fits
(response is MCX)
800

600

400

200
Residual

-200

-400

-600

-800
1800 2000 2200 2400 2600 2800 3000 3200
Fitted Value

Figure 2-15 Residual for Regression between MCXCODMEX and Index of industrial
Production

Result

The regression equation is

MCX = 96 + 8.26 IIP
R2 = 53.3%
Hypothesis test

• Testing first hypothesis

H0 : R2 = 0
As R2 = 53.3 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%
As R2 = 53.3 % > 50 % so the test fail to accept the Hypothesises, IIP is strongly
associated with Commodity Price
Now from the Equation it appears that value of Commodity Price is positive related with
Money Supply, but as R2 value is just low (>50 %) ,so the relationship is strong, but it is
almost at par to Money Supply.

37 | P a g e
 Regression of MCXCOMDEX along with IIP

Versus Fits
(response is MCX)
1000

500
Residual

-500

-1000
2200 2300 2400 2500 2600
Fitted Value

Figure 2-16 Residual for Regression between MCXCODMEX and Prime Lending Rate

Result

The regression equation is

MCX = 3041 – 59.1 Prime lending rate
R2 = 4.2%
Hypothesis test

• Testing first hypothesis

H0 : R2 = 0
As R2 = 4.2 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%
As R2 = 4.2 % < 50 % so the test fail to reject the Hypothesises, interest rate is loosely
associated with Commodity Price
Now from the Equation it appears that value of Commodity Price is negatively related
with Interest rate, but as R2 value is just low (<50 %) ,so the relationship is not very
strong and it is almost close to zero.

38 | P a g e
 Multiple Regression of Commodity Price with all Monetary Factors
In order to develop the Complete Model, all Monetary Indicators have been included
and a Multiple Regression has been done

Versus Fits
(response is MCX)
750

500

250
Residual

-250

-500
1600 1800 2000 2200 2400 2600 2800 3000 3200
Fitted Value

Figure 2-17 Residual for regression of MCXCODMEX with CPI, Money Supply, IIP, Interest
rate, Ex Rate

Result

The regression equation is

MCX = 4906 - 5.21 CPI + 0.00457 Money Supply + 0.79 IIP
- 30.9 Prime Lending Rate - 62.3 Ex
2
R = 69.6%
Hypothesis test

• Testing first hypothesis

H0 : R2 = 0
As R2 = 69.6 # 0 so the test fail to accept the hypothesis, so
• Testing second hypothesis
H0 : R2 < 50%

39 | P a g e
 As R2 = 69.6 % > 50 % so the test fail to accept the Hypothesises, all five monetary
indicator are not loosely associated with Commodity Price
Now from the Equation it appears that value of Commodity Price is strongly related with
all monetary indicators. This is the final complete Model and R2 justify that model is
robust enough to predict commodity price with fluctuation in any of the given 5 monetary
indicator

MCX Sub indices – Metal, Energy, Agri

Figure 2-18 Movement of Sub-indices of MCX along with Time

In order to see the effect of monetary indicator on sub-indices, we first evaluate that whether sub
index is highly related MCXCODMEX. If they are highly correlated than Monetary Indicator has
same results. So Correlation of Individual Sub index with MCXCODMEX has been deduced

Correlations: MCXCODMEX, SUB INDICES

Pearson correlation of MCX and METAL = 0.895

Pearson correlation of MCX and ENERGY = 0.762
Pearson correlation of MCX and AGRI = 0.609

40 | P a g e
Now only MCX AGRI is not strongly correlated compare to others, so a separate Multiple
regression has been done MCX Agri data

Multiple Regression of MCX Agri with all Monetary Factors

Versus Fits
(response is MCX)
300

200

100
Residual

-100

-200

-300

-400
1500 1750 2000 2250 2500
Fitted Value

Figure 2-19 Residual for regression of MCX Agri with CPI, Money Supply, IIP, Interest rate,
Ex Rate

Result

The regression equation is

MCX = 2326 + 3.25 CPI + 0.000789 Money Supply - 0.80 IIP
- 58.9 Prime Lending Rate - 33.9 Ex
2
R = 84.8%

R2 has been jumped from 69% to 84% it shows , that in case of choosing only agri commodity ,
all monetary indicator show better association. It implies that in case of Metal and Energy there
are others factors who determine the price also e.g. demand-supply with effective weight.

41 | P a g e
 NCDEX

Figure 2-20 Movement of Dhanya (NCDEX) index along with time

NCDEX has only one index “Dhnya” and that is only based on agri product .So first it need to be
check whether the Dhanya Index and MCX are correlated or not ,only in that case effect of
Monetary Indicator can be drawn for Dhanya Index

Correlation between Dhanya (NCDEX) and MCX Agri

Results

Pearson correlation of Dhanya and MCX AGRI = -0.624

The correlation coefficient of -.624 shows that both indices are negatively correlated, it means in
spite of the fact that both are agri index, they contain such product whose price very oppositely
with Monetary indicator fluctuation. It can be also proved by looking at both indices component,
while NCDEX Index is based on pulses while MCX Agri is based on perishable product which
creates the difference. Also even both of them move oppositely but they have a significant
relation

42 | P a g e
 Chapter 3

Interpretation

Exchange Rate
The exchange rate shows very minimal effect on commodity price, it may be possible that
the commodity price is based on domestic market, so impact of exchange rate is quite low.
Exchange rate may have significant effect, if international market will be involve in terms of
import and export. Even in overall multiple regression exchange rates doesn’t show a significant
effect .But there is unique observation while looking the equation as stand-alone it provide
positive effect while for complete model it shows a negative impact. So we can infer that as a
stand-alone factor Exchange rate increase the commodity price with increasing value ,but for a
complete model it reduce the commodity price. It also implies that Exchange rate have some
association with other monetary indicators.

Consumer Price Index (CPI)

In case of Consumer Price Index the effect is positive as like exchange rate , but its
significance is quite good. Consumer Price Index is a major indicator for inflation measures, so
we can deduce that increasing inflation as stand-alone factor the Commodity price will
increase ,but in all practical purpose it is impossible to do, So in complete model it show a
negative effect just like exchange rate. In overall monetary policy development inflation is a
major indicator and if government failed to curb it, it will lead to a very high commodity price.

Money Supply (M)

Money Supply justify policy’s liquidity aspect ,it can be observe from equation that
increase in money supply will lead high inflation and that lead to high commodity price, but in
complete model it behaves differently compare to CPI. It act as a positive agent and increase
Commodity price as it itself grow. It might possible that other Monetary factor dominate CPI ,so
as it behaves oppositely in complete model, but Money Supply contain larger effect ,so it cannot
be dominated by other indicators.

43 | P a g e
 Production Index (IIP)
The most unexpected impact is of Production Index, it not only impact positively ,it also
have a significant influence with regression coefficient value above 50%.Just like the case of
Money Supply it is also a dominating factor ,which affect the Commodity Price even in the
presence of other Monetary Indicators.

Interest Rate
This the most important and difficult factor to analyse. First of all it is very difficult to
calculate the real interest rate in the absence of exact inflation value. Still forms stand-alone
model it can infer that interest rate negatively affect the commodity price ,which has been proved
several time earlier. The important part is to judge the effect with other indicators and we can see
that interest rate prove to be a dominating factor even in complete model and act as a negative
factor for commodity price.

Recommendation

• From all the above statistical analyse it can be infer that the model for commodity price is
more accurate as more & more indicator incorporated for regression. It happens because
commodity price is driven by a number of qualitative and quantitative factors. All
possible quantitate factor need to be incorporated in the model and once the model is
developed it need to support the qualitative aspect also.
• Determination of any monetary policy cannot be based on single commodity index.
Ideally statistical analysis need to be perform on sector specific indices e.g. Metal,
Energy, Agri, Oil etc.in Appendix the same statistical treatment has been given for
NCDEX index “Dhanya”.
• The current model gives equal weightage to all the factors, but for all practical purpose
these factor need to be weighted as per their characteristics. Analyst Need to perform a
separate an analysis for short term and long term interest rate because they depicts
different policy requirement.
• The current analysis is only based on Commodity spot price, commodity exchange
contain a large database on future market, So any implementation of policy is more useful
if it cover both spot and future price fluctuation and their driving factors.

44 | P a g e
 • The price discovery process should not be left to just a handful of traders in
asymmetrically informed or ill informed, segmented markets. Rather, the best price
discovery comes when a large number of various categories of market players with
a wide range of objectives and interests converge on an organized futures platform.
Such a platform and the Multi Commodity Exchange of India Ltd. (MCX) is one
ensures that all relevant information is absorbed in the price formation process, and the
“right price” is discovered. The more efficient the discovered prices on a futures
platform is, the more effective are the business and policy decisions that are taken
based on these prices.

At this juncture when the Indian markets are on their way to the heights achieved by the global
benchmark markets, it is essential that they are allowed to have the right mix of participants and
products to have the necessary liquidity depth and width. Corporates and physical market players
in India are gradually realising the importance and need to participate on commodity exchanges.
An increased participation of such players will go a long way in streamlining commodity trading
in India by bringing in relevant information about the fundamentals into the markets and, thus,
making the price discovery process more efficient. Besides, this will also help corporate best
practices percolate into the markets to fine-tune their functioning and efficiency. Given the
current trend of globalisation of economies, competitiveness remains one of the most defining
factors for developing economies. And this not only means having competitive manufacturing
and services sectors but also necessitates promotion of markets to make them globally-
competitive.

Conclusion
Theoretical explanation that monetary policy makers need to consider commodity price as an
indicator is easy task but actual implementation is quite hard. But not everyone would consider it
45 | P a g e
obvious that an index of agricultural and mineral commodity prices belongs on a useful list of
variables to reveal current monetary conditions, alongside interest rates, the exchange rate,
money supply and Production rate. The conventional practice is to throw the volatile “food and
energy” sector out of the price indices, concentrating instead on the core CPI if one wants a good
indicator of likely future inflation. It is certainly true that if one is looking for the single standard
statistic that best predicts future inflation, the core CPI will do better than the adjusted CPI. But
true monetary control comes by a free to look at lots of information. For example considering are
agricultural and mineral prices on the list of variables have significant impact. Our perspective
places commodity prices on a plane with central banks paying attention to housing prices or the
stock market.
The theory and empirical results reported in dissertation report suggest that the
commodity prices belong on the list of monetary condition indicators. Real commodity prices
reflect monetary ease, more specifically real interest rates, among other factors. We can never be
sure what the real interest rate is, because we do not directly observe expected inflation. Thus it
is useful to have additional data that can be expected to reflect real interest rates.

The main empirical results of dissertation report are quite insightful. At a global level, several
literature support the conjecture that monetary aggregates may convey some useful information
on variables such as commodity prices which matter for aggregate demand and hence inflation.
Moreover, report identifies a negative relation between the world interest rate and commodity
prices as proposed by various other researchers. Thus, we conclude that even in India liquidity,
Production Development and the interest rate are useful indicators of commodity price inflation
and of a more generally defined inflationary pressure at a global level. Given that the interest
rate is inversely reacting to output and commodity prices, but it might be not showing exact
picture as the interest rate doesn’t seem to have been adjusted enough in the long-run to account
for commodity price inflation and output growth on a global scale. Therefore we would like to
argue that Indian economic liquidity merits some attention in the same way as the worldwide
level of interest rates received in the recent hot debate about the world savings and liquidity glut
as the main drivers of the current financial crisis, if not possibly more. Expressed on a more
technical level, this paper has analyzed the relationship among money, interest rates and
commodity prices for Indian economy in theoretical nature. At the OECD level, report gather

46 | P a g e
further support of the conjecture that monetary aggregates may convey some useful information
about the future development of commodity prices which matter for aggregate demand and hence
consumer price inflation. Dissertation’s empirical results appear to be overall robust since they
pass most of the hypothesis for strong relationship.
One further advantage might be the more timely availability of commodity price data
relative to those on overall monetary indicator. Exchange like MCX and NCDEX are new to
Indian economy, so data is available for year 2005 only, but still the statistical result show
effective model has been established.A holistic study of the monetary policy is must to
understand the formulation of Commodity Price. The interesting Part is that, not only Monetary
Indicator affect the Commodity Price, commodity price also influence the monetary policy.
Several studs i.e. have been done to analyses these and by digging more & more economist will
be able to develop a policy system which will be robust in nature and included all aspect of
economic growth driver.

Bibliography

1. Hanke and Wichern(2009). “Busines Forecasting” ,8th Edition ,PHI publication

house,New Delhi.

47 | P a g e
 2. Pindyank(2009). “Macroeconomic Theroy”, 5th Edition ,TATA Mcgraw Hill
Publication,New Delhi.

3. Awokuse, T.O. and Yang, J.(2003). The Information Role of Commodity Prices in
Formulating Monetary Policy: A Reexamination. Economics Letters.

4. Cody, B.J. and Mills, L.O.(1991). The Role of Commodity Prices in Formulating
Monetary Policy. Review of Economics and Statistics.

5. Garner, A.C.(1989). Commodity Prices: Policy Target or Information Variables?

Journal of Money, Credit and Banking.

6. Hua, P.(1998). On Primary Commodity Prices: The Impact of Macroeconomic/

Monetary Shocks. Journal of Policy Modeling.

7. Marquis, M.H. and Cunningham, S.R.(1990). Is There a Role of Commodity Prices in

the Design of Monetary Policy? Some Empirical Evidence. Southern Economic
Journal.

8. Sephton, P. S.(1991). Commodity Prices: Policy Target or Information Variables? A

Comment. Journal of Money, Credit and Banking.

9. Toda, H.Y. and Yamamoto (1995). Statistical Inference in Vector Auto regressions

10. Alquist, R. and L. Kilian (2010), .What do we learn from the price of crude oil futures?
Journal of Applied Econometrics, forthcoming.

11. Anzuini, A., P. Pagano and M. Pisani (2007), .Oil supply news in a VAR: Information
from financial markets., Temi di discussione No. 632, Banca d.Italia.

12. Barsky, R.B. and L. Kilian (2002) .Do we really know that oil caused great stag.ation? A
monetary alternative., NBER Macroeconomics Annual.

13. Deffeyes, Kenneth (2005) Beyond Oil: The View from Hubbert’s Peak (Hill and Wang).

14. Dornbusch, Rudiger (1976) "Expectations and Exchange Rate Dynamics" Journal of
Political Economy.

15. Frankel, Jeffrey (1995) "Expectations and Commodity Price Dynamics: The
Overshooting Model," Amer. J. of Agric. Reprinted in Frankel, Financial Markets and
Monetary Policy, MIT Press.

16. "Commodity Prices and Money(1984): Lessons from International Finance," American J.
of Agr. Economics 66.

48 | P a g e
 17. Frankel, Jeffrey and Gikas Hardouvelis(1985): "Commodity Prices, Money Surprises,
and Fed Credibility," Journal of Money, Credit and Banking 17.

18. EBSCO Database

19. Prowess Database

20. www.capitaline.com

21. www.indiastast.com

22. www.rbi.com

23. www.wikipedia.com

24. www.mcxindia.com

25. www.ncdex.com

26. www.wikipedia.com

27. www.moneycontrol.com

28. www.oecd.org

49 | P a g e
Appendix

Analysis of NCDX (“Ganya”)

Minitab Analysis of same monetary indicator with other Commodity Index

Versus Fits
(response is Dhnaya Monthly)
250

200

150
Residual

100

50

-50

-100
600 700 800 900 1000 1100
Fitted Value

The regression equation is

Dhnaya Monthly = - 8.1 + 0.00129 Money Supply

R-Sq = 87.1%

50 | P a g e
 Versus Fits
(response is Dhnaya Monthly)
300

200

100
Residual

-100

-200

600 700 800 900 1000 1100 1200

Fitted Value

The regression equation is

Dhnaya Monthly = - 472 + 4.50 IIP
R-Sq = 62.5%

Versus Fits
(response is Dhnaya Monthly)

300

200

100
Residual

-100

-200

700 800 900 1000 1100

Fitted Value

The regression equation is

Dhnaya Monthly = 1675 - 69.9 Prime Lending Rate
R-Sq = 44.8%

51 | P a g e
 Versus Fits
(response is Dhnaya Monthly)
300

200

100
Residual

-100

-200

-300
700 750 800 850 900 950 1000
Fitted Value

The regression equation is

Dhnaya Monthly = - 300 + 25.1 Ex

R-Sq = 27.1%

Versus Fits
(response is Dhnaya Monthly)
200

150

100
Residual

50

-50

-100
600 700 800 900 1000 1100
Fitted Value

52 | P a g e
TRENDS IN INDEX OF INDUSTRIAL PRODUCTION
TRENDS IN INDEX OF INDUSTRIAL PRODUCTION
(Base : 1993-94=100)
Sector Mining & Manufacturing Electricity General
Quarrying
Weight 10.473 79.358 10.169 100
Period Index Growth Index Growth Index Growth Index Growth
Rate Rate Rate Rate
(per (per (per (per
cent) cent) cent) cent)
1 2 3 4 5 6 7 8 9
2001-02 131.9 1.3 172.7 2.9 159.2 3.1 167 2.8
-3.7 -86.6 -10.8 -100
2002-03 139.6 5.8 183.1 6 164.3 3.2 176.6 5.8
-8.4 -86 -5.4 -100
2003-04 146.9 5.3 196.6 7.4 172.6 5 189 7
-6.2 -86.4 -6.8 -100
2004-05 153.4 4.4 214.6 9.2 181.5 5.2 204.8 8.4
-4.3 -90.4 -5.7 -100
2005-06 154.8 0.9 234.1 9.1 190.9 5.2 221.4 8.1
P
-0.9 -93.2 -5.8 -100

Monthly Average Data for All Monetary Indicators

Mone
y Prime
Supp Lending Exchan
MCX CPI ly IIP Rate ge Rate
3220. 87853
38 570 2 331.5 8.5 45.31
3239. 87189
99 566 7 325.7 8.5 44.81
3085. 84080
39 562 6 323 8.5 46.04
2922. 82621
42 557 3 346.3 8 44.54
2798. 82301
73 554 3 325.9 8 44.92
2728. 81452
37 547 6 323.9 8 47.08
2726. 83244
87 540 9 323.9 12 46.46
2644. 82515
65 538 2 373.8 12 46.6
2633. 79873
75 536 5 331.1 12 46.45

53 | P a g e
 2680. 76803
67 538 3 331.2 12 44.44
2645. 75538
29 542 6 334.3 12 45.14
2637. 74214
87 538 5 299.8 12 46.23
2528. 73826
58 532 6 289.7 12 46.37
2724. 72369
37 522 1 302 12 46.68
2672. 71287
75 515 0 292.8 12 46.48
2606. 69027
5 508 9 290.8 12 46.96
2450. 67912
8 499 9 291.6 12 48.04
2476. 67596
22 484 3 280.3 12 48.88
2329. 69036
54 475 2 269.3 12.25 48.16
2343. 69487
4 468 0 305.9 12.25 47.87
2235. 68712
73 463 7 276.8 12.25 47.29
2017. 66545
91 462 0 284.8 12.5 50.22
1978. 65012
07 461 0 284 12.5 50.95
1881. 63380
97 459 4 267.6 12.5 50.73
1726. 62655
82 460 0 262.9 13.25 49.02
1645. 61898
34 459 7 276.2 13.5 48.45
1859. 61086
17 455 0 264.7 14 49.84
2051. 58614
08 450 6 271.3 14 49.25
2762. 58670
67 442 8 269.2 14 46.94
2734. 60288
58 434 5 274.6 13.25 43.79
2889. 60418
35 431 2 266.3 12.75 42.49
3182. 60259
65 429 8 304.9 12.75 42.95
3116. 58931
3 423 9 276.2 12.75 42.59
2825. 417 56841 281.9 12.75 40.46

54 | P a g e
 38 0
2722. 55420
13 413 4 284.7 13.25 39.97
2667. 54419
03 413 6 261 13.25 39.92
2405. 53401
78 414 0 262.6 13.25 39.39
2382. 52511
98 413 3 260.5 13.25 39.41
2347. 50606
92 410 8 260.3 13.25 39.67
2319. 48638
62 408 7 255 13.25 39.32
2210. 48655
47 404 4 255.3 13.25 39.74
2056. 49563
68 399 2 263.1 13.25 40.96
2189. 50114
06 395 3 250.7 13.25 40.44
2121. 50392
38 394 2 289.1 13.25 40.75
2106. 50261
96 392 2 252.2 13.25 40.73
2133. 48280
03 392 5 265.5 12.5 41.29
2132. 48243
57 391 9 263.7 12.5 43.59
2162. 47198
31 390 1 248.8 12 44.31
1996. 46162
96 390 4 234 11.5 44.17
2139. 45218
4 386 0 243.5 11.5 44.23
2123. 45166
88 380 5 234.8 11.5 44.76
2087. 43403
52 375 7 235.5 11.5 45.02
2323. 43785
48 372 6 234.4 11.5 45.96
2280. 43518
04 370 6 237.9 11.25 46.55
2162. 43691
83 365 6 225.2 11.25 46.51
2308. 43616
34 360 2 251.9 11.25 46.08
2189. 43163
72 358 9 227.3 10.75 46.43
1917. 41311
12 357 9 237.9 10.75 44.97

55 | P a g e
 1838. 41109
77 357 2 232.5 10.75 44.61
1938. 40353
63 358 4 214.8 10.75 44.44
1783. 39465
86 360 2 223.9 10.75 44.07
1742. 38774
92 356 7 217.4 10.75 45.07
1763. 38033
7 354 6 212.9 10.75 45.94
1670. 36452
44 352 8 208.1 10.75 45.11
1607. 37146
15 350 9 213.6 10.75 43.99
1636. 37098
67 345 8 213 10.75 44.04

56 | P a g e