IS MONEY DEMAND IN

INDIA STABLE? AN
EMPIRICAL STUDY

SUBMITTED BY:

Arpita Katiyar ECO2216

Jiya Bharati ECO2225

Kritika Chhapolia ECO2226

SUBMITTED TO:

Dr. Ranjit Pattnaik

INTRODUCTION:

Money demand stability is a topic that has garnered significant attention among economists

and policymakers for decades. The concept refers to the predictability of the demand for

money over time, and its ability to remain relatively stable in response to changes in various

economic factors. The stability of money demand is an important aspect of macroeconomic

stability as it affects monetary policy decisions, interest rates, and inflation. As such,

understanding the factors that contribute to money demand stability is crucial for formulating

effective economic policies and achieving macroeconomic stability.

One explanation of the instability in money demand has been attributed to the financial

innovations and institutional changes which have been able to explain the instability of

money demand in the US economy to a very large extent. Another explanation emerges in the

form of unexpected changes in the income velocity of money. (Jonung 2018) Instability of

money demand could also be explained by changes in the money stock held as the money

stock held may not correspond to the money stock actually desired. (Andersen 1985)

To study the stability of money demand, it is essential to understand the best way to define

the money demand functions. Various theoretical models have emerged to explain the same,

some of which will be discussed in this paper, along with an analysis of the stability of

money demand in the Indian context over different time periods between 1997 and 2022. Part

I of the paper deals with an explanation of what money demand and its stability entails. Part

II talks about four theories of money demand developed by economists. Part III is a rundown
of the data and the methodology used in the forthcoming analysis. Part IV is about the results

of a regression analysis on Indian money demand data and Part V contains the concluding

remarks as well as policy implications based on the findings.

PART I

Money demand is a fundamental concept in macroeconomics that seeks to capture the

amount of money that firms and individuals require for transactions and investment purposes.

Money demand could include only cash which is the most liquid asset or other related liquid

assets like demand deposits. Money demand is not directly observable in the economy and

money supply is used as a representative of the demand for money in an economy because

this supply is exogenously fixed by central banks in response to the demand experienced in

the economy. Therefore, many measures of money supply have emerged over the years and

in India, new monetary aggregates have emerged in the recent past such that the definitions of

narrow and broad money have undergone a significant change.(Sriram 1999)

Given the definition of money demand, it is important to note what stability of money

demand actually entails and lay down the conditions that would prove that money demand is

stable. (Ailin Xie 2017) Therefore, three conditions must be satisfied for money demand to be

stable which are:

a. The demand for money should be predictable in a statistical sense such that

forecasting can be done satisfactorily making money demand an important input for

monetary policy

b. The explanatory variables in the money demand function should be limited

c. The explanatory variables in the function should be closely linked to real economic

activity
Many economists have tried to measure the stability of money demand in western economies

and noteworthy among them is Goldfeld who tried to forecast outside the sample period in

order to gain a sense of how stable money demand is. His analysis of the US money demand

yielded that money demand was stable until 1975 in the US economy after which certain

changes had to be made to the analysis given the changing nature of the economy as well as

the financial innovations constantly taking place and disrupting the market.

PART II

Quantity Theory of Money

The classical quantity theory of money has two major approaches: Fisher’s equation of

exchange and the cash balance approach of Cambridge economists. Both the theories are two

sides of the same coin and seek to define money demand as represented by the amount of

money being used in the economy. (Mandereau 1934)

In Fisher’s transaction approach to the quantity theory of money, money demand is defined

by the equation:

MV=PT

Where the important component is V which represents the velocity of money circulation in

the economy, i.e. the number of times money exchanges hands in the economy. The product

of the velocity of money circulation and the amount of money in circulation is said to be

equivalent to the total value of the transactions taking place in the economy.

In the cash balance approach, money demand is explained as a certain proportion of the

nominal GDP:

M= kPY
‘k’ is included to explain the fact that money is demanded for purposes other than

transactions such as for future transactions. However, ‘k’ does not have a deterministic value

like 1/v in the Fisher equation. Hence, the ‘k’ in the Cambridge equation is an unstable value

which makes the Cambridge equation unstable from its very conception and ‘k’ has also not

been clearly defined by Cambridge economists, which is a major drawback of the cash

balance approach.

Baumol-Tobin transactions demand model

This theory of money demand is based on the idea that there is always a tradeoff between the

cost of holding money in terms of the interest foregone and the inconvenience of holding

small amounts of money. Minimizing the cost and considering the average money holdings,

the money demand can be represented as:

Where,

tc= cost of moving between money and bonds

Y= income

i= interest rate

Therefore, according to this model, money demand increases with increase in income and

increase in the transaction costs, while money demand decreases with increase in the interest

rate. Moreover, the income elasticity of money demand and the interest elasticity of money

demand are given as ½ and -½ respectively. (Encarnación 1973)

Friedman’s restatement of the quantity theory of money

According to Friedman, money matters and if money is held for its use as purchasing power

for other goods, money can also be considered as a good in its real and nominal form.

Therefore, money and alternative assets like bonds and equity are incorporated into an

individual’s utility function. Moreover, inflation causes the value of money to erode and

therefore, acts as the opportunity cost of holding real balances as against commodities. Given

Friedman’s monetarism, real money demand can be expressed as,

M= f(ri, ç, w, HW/NHW)

Where,

ri= real return on ith asset

Π= expected rate of inflation

w= wealth in real terms

HW/NHW= ratio of human to non-human wealth

Therefore, according to Friedman’s theory, real money demand depends on the real return on

alternative assets like bonds and equity, the expected rate of inflation, wealth in real terms

and the ratio of human to non-human wealth.(Georgiev 2007)

Buffer stock model

This model is based on the idea that people have a desire to maintain a certain level of

financial security, or "buffer stock," to provide a cushion against unexpected changes in

income or expenses. The buffer stock model of money demand suggests that people will

adjust their demand for money based on changes in their income level or variability. For

example, if a person's income increases, they may choose to hold a larger buffer stock of

money to maintain their desired level of financial security. Similarly, if a person's income

becomes more variable, they may also choose to hold a larger buffer stock of money to

protect against income fluctuations.(Mizen 1994)

The buffer stock model of money demand also suggests that changes in interest rates may

have a limited impact on the demand for money. While higher interest rates may increase the

opportunity cost of holding money (i.e., the cost of forgoing interest earnings on other assets),

they may not necessarily reduce the demand for money if people are motivated by a desire to

maintain a certain level of financial security.

The buffer stock model of money demand is based on two key assumptions:

1)People prefer to have a certain level of financial security to buffer against unexpected

expenses or changes in income.

2)The level of financial security that people prefer is determined by their income level and

the variability of their income.

The buffer stock model allows money demand to fluctuate within an upper and lower limit

known as thresholds or around a long run desired money demand. The two versions of buffer

stock models are: one, where a policy decision is made to limit money demand within a set

minimum and maximum and therefore, when money demand moves past the limits, it is

corrected by policy; two, objective or smoothing models.

PART III

METHODOLOGY: -

The secondary Data has been extracted from Reserve Bank of India - Handbook of Statistics on
Indian Economy and India, GDP, Quarterly - Economic Data Series | FRED | St. Louis Fed , for
estimation of multiple linear regression , ordinary least squares (OLS) has been used with the help of
STATA Software. Variables which we have considered are Broad Money(M3), Consumer Price Index
(CPI) as a proxy for inflation, Weighted Average Call Rate (WACR) as proxy for interest rate and Real
GDP as proxy for Income level. For this analysis we have considered yearly data from 1997-2011,
2012-2016, 2017-2021.

DATA ANALYSIS

Money Demand=f {WACR, GDP}

As a MLRM model, money demand function can be written as:

Money Demand=b0+b1*WACR+b2*GDP

Here money demand is dependent variable and weighted average call rate (WACR) and GDP are
explanatory variable (independent). where b0 is our intercept term and b1 and b2 are our partial
regression coefficients showing change in money demand with one unit change in either WACR or
GDP while holding the other constant.
INTERPRETATION: -

Coefficients p-value
Intercept -6.33826 9.05e-15
Wacr -0.09725 0.136
Gdp 0.75754 2e-16

Money demand=-6.33826+(-0.09725*Wacr)+0.75754*Gdp

The above equation is the estimated equation for Money Demand.

Interpretation on different intervals.

In 1st time period (1997-2011), WACR is statistically insignificant i.e. it is not feasible to take it as an
independent variable. GDP as an independent variable is statistically significant.

Adjusted r squared value= 0.97988

Coefficients p-value
Intercept -7.37628 3.99E-09
Wacr -0.17312 0.032515
Gdp 0.850635 5.78E-11

As p-value for WACR is less than 0.05 (level of significance) , we reject our null hypothesis that
WACR is statistically significant.

GDP is statistically significant as the p-value for it is less than 0.05.

2nd Time Period

In 2nd time period (2012-2016) we have found that wacr is statistically insignificant i.e. it is also not
feasible to take it as an independent variable. GDP as an independent variable is statistically
significant.

Adjusted r squared value= 0.923787

Coefficients p-value
Intercept -0.8586 0.817568
Wacr -0.13179 0.573338
 Gdp 0.315686 0.338264

As p-value for WACR is more than 0.05 (level of significance) , we accept our null hypothesis that
WACR is statistically insiginificant.

GDP is also statistically insignificant as the p-value for it is more than 0.05.

3rd Time Period

In 3rd time period (2017-2021) we have found that both WACR and GDP are statistically insignificant
i.e. it is not feasible to take it as an independent variable

Adjusted r squared value= 0.956609

Coefficients p-value
Intercept 1.582 0.506194
Wacr -0.21272 0.092966
Gdp 0.124645 0.510361

As p-value for WACR and GDP is more than 0.05 (level of significance) , we accept our null
hypothesis that WACR and GDP are statistically insignificant.