




































 

 

 
1 

© 2020 Conscientia Beam. All Rights Reserved. 

 

EVALUATING THE MODEL OF DEMAND FOR MONEY IN NIGERIA   

 

           

Fasipe, Temidayo B.1+ 

 Wasiu A. Yusuf2 

 

1Central Bank of Nigeria (HQTR), Abuja, Nigeria. 

 
2Nile University of Nigeria, FCT Abuja, Nigeria. 

 
  

(+ Corresponding author) 

 ABSTRACT 
 
Article History 
Received: 13 November 2019 
Revised: 19 December 2019 
Accepted: 23 January 2020 
Published: 16 March 2020 
 

Keywords 
Broad money 
Monetary policy 
Narrow money 
Treasury bill 
Saving-deposit rate. 

 
JEL Classification: 
E41; C42. 
 

 
In Nigeria, demand for money is frequently affected by factors that regularly experience 
shocks in the economy. Therefore, regular adjustment and reforms are done to 
monetary policy which creates a lot of uncertainties in the market. This paper therefore 
examined the demand for money (narrow money, M1) in Nigeria, using quarterly time 
series data from 2006 to 2018, the study attempted multiple OLS regression analysis 
and ARDL. The result found out that money demand function cannot be appropriately 
estimated by OLS estimation technique due to the presence of the lagged value of both 
the dependent and independent variables. Although, the no long run relationship 
among the variables but the result indicates that M1 is largely influenced by inflation, 
exchange rate, MPR (Monetary Policy Rate), and savings as well as real GDP to some 
extent; particularly in the short run. It was observed from the analysis that economic 
units in Nigeria are shedding more of cash assets (Naira) as inflation increases while 
stocking up on foreign cash and assets (dollar and foreign denominated assets) as 
shown by the positive-related exchange rate.  
 
 

Contribution/Originality: This study contributes to the existing literature by examining the demand for 

money (narrow money, M1) in Nigeria and using quarterly time series data from 2006 to 2018, the study attempted 

multiple OLS regression analysis and ARDL. 

 

1. INTRODUCTION 

1.1. Background to the Study 

The subject of demand for money has always generated keen interest among research and theoretical 

economists in both the developed and developing countries. This high interest stems from the facts that demand for 

money plays a major role in empirical macroeconomics analysis and policy decision formulations. The interest has, 

however, heightened in recent years, and may attributable to concerns among central banks and researchers on the 

economic impact of the movement towards flexible hang rate regime, globalization of capital markets, ongoing 

domestic financial liberalization, advancement in time series econometrics, and some other country-specific issues. 

In Nigeria, the monetary authorities acknowledge that a sound specification and estimation of money demand 

function is essential in the selection of instruments and targets, so also is the fact that the transmission mechanism 

of monetary policy depends on how correct its specification and estimation. The Central Bank of Nigeria is charged 

with promoting monetary stability and a sound financial system in Nigeria (Central Bank of Nigeria (CBN), 2018; 

Research Department Central Bank of Nigeria, 2016). Therefore, in pursuit of its mandate, the monetary policy 

Financial Risk and Management Reviews 
2020 Vol. 6, No. 1, pp. 1-13. 
ISSN(e): 2411-6408 
ISSN(p): 2412-3404 
DOI: 10.18488/journal.89.2020.61.1.13 
© 2020 Conscientia Beam. All Rights Reserved. 

 
 
 

 
 
 

 

 
 
 
 

https://orcid.org/0000-0003-0461-3253
https://www.doi.org/10.18488/journal.89.2020.61.1.13


Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
2 

© 2020 Conscientia Beam. All Rights Reserved. 

measures adopted by the Central Bank have direct and/or indirect effects on money demand. Deliberate policy-

driven changes in monetary policy tools such as Monetary Policy Rate (MPR), Cash Reserve Requirement (CRR) 

and other reserve requirements induce fluctuations in interest rates, exchange rates and GDP, which in turn bring 

about changes in money demand in the economy. However, despite the enormity of the importance of monetary 

policy instruments as it affect demand for money; there is dearth of literature on the subject matter. Therefore, 

there is the need to constantly assess the model of demand for money with a view to determine its stability over 

time as well as its speed of adjustment to monetary policy changes. The study also aimed at creating a background 

to review the effectiveness of the Central Bank’s monetary policies in the context of the overall macroeconomic 

stability; evaluate the determinants of money demand and their impacts on demand for money in Nigeria. In 

addition, the study will also provide a basis to advance recommendation in relation to monetary policy framework 

in the country. 

 

2. LITERATURE REVIEW 

2.1. Theoretical Review 

Many of the theories on demand for money specify different explanatory variables ranging from inflation, 

interest rate, transactions, utility, wealth, etc. One important feature of these theories is that they share almost same 

determinant variables; however, they differ in the specific role and significance attached to each variable. In general, 

theoretical models of money demand have underpinned and motivated a great deal of empirical work. 

The quantity theory examined the relationship between the total quantity of money M and the total amount of 

spending on final goods and services produced in the economy P×Y, where P is the price level and Y is aggregate 

output (Busari, 2005). In its basic form, the theory specifies that  

       (1) 

Where M is the quantity of money, V is the velocity of circulation; P is the price levels and T the volume of 

transactions. Because V cannot be measured empirically, Equation 1 becomes a mere identity. If we assume T moves 

approximately in tandem with the real GDP (Y), (1) leads to the standard form of quantity theory (see Equation 2: 

       (2) 

Which is still an identity and not yet a theory of money demand. A variant of the theory re-specifies demand for 

money as shown in Equation 3 and 4:  

       (3) 

 and then, 

        (4) 

Equation 4 shows that the demand for money is proportional to the amount of real transactions, represented by 

the real GDP; and is proportional to the price level. It can also be interpreted in a way that the demand for money is 

actually a demand for a real quantity of money as in Equation 5, where MD/P represents the demand for real 

quantity of money: 

          (5)  



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
3 

© 2020 Conscientia Beam. All Rights Reserved. 

One important difference between the cash balance approach and subsequent theories of the demand are the 

lack of an interest rate variable in the former. Moreover, the fundamental assumption of the quantity theory as 

regards stable velocity of money lacks empirical support.  

The Liquidity Preference Theory emphasized the role and importance of interest rates. Keynes (1930) in his 

book titled “The General Theory of Employment, Interest and Money” introduced three motives for holding money 

(Essien, Onwioduokit, & Osho, 1996): (transaction motive, precautionary motive, speculation motive). The 

speculative demand for money is an inverse function of the rate of interest on bonds, . Together with a 

transaction demand for money,  and a precautionary demand for money, both being a direct function of income, Y, 

the total demand for money MD (shown in Equation 6) now becomes: 

           (6) 

However, a completely different approach was chosen by Baumol (1952) and Tobin (1956). The model seeks to 

determine the optimum level of demand for money considering the trade-off between one the opportunity cost of 

holding money in term of interest on securities foregone, and two, the transaction cost of converting money into 

securities. 

That is, the total cost for the payments services (PS) is as given in Equation 7: 

 

=                   (7) 

Where TC is the transaction cost of conversion, OC is the opportunity costs of holding money, n the number of 

transactions in a given period, i the interest rate of securities and M equals PY/2n (that is, the average money 

holdings in the period are determined by monthly income, PY, and the number of transactions).The minimum cost 

is obtained on differentiating PS with respect to M (see Equation 8): 

 

Which gives; 

          (8) 

Taking the nominal transaction costs can be regarded as the product of the real transaction costs (CR) and the 

price level (P), i.e. C = CR.P, thus Equation 9 leads to the optimum  real stock, which is identical with the optimum 

real money demand. 

                        (9) 

Using m for M/P and transforming into a natural logarithmic form, gives Equation 10:  

                   (10) 

The conclusions derivable form the Baumol–Tobin money demand function is that the demand for non-interest 

bearing money in the sense of M1 depends positively on real income, negatively on interest rate, and positively on 

real transaction costs. 

In addition, Friedman (1956) argues that the demand for money should be treated in the same way as the 

demand for goods or services. He defines the total wealth of an individual as sum of five components: money, bonds, 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
4 

© 2020 Conscientia Beam. All Rights Reserved. 

shares, real assets and human capital. Therefore, a theory of demand for money would require information on each 

of the components and on the individual returns therefrom (Friedman, 1956). He stated the money demand function 

as Equation 11. 

                      (11) 

Where w is the fraction of wealth in non-human form,  are the expected nominal rates of returns on 

money, bonds and physical asset; and    stands for other variables attached to services of money.  

 

2.2. Empirical Review: Models of Money Demand Function 

Dou (2018) observed that money demand and its stability have a great impact on the economy of a country. 

Since China’s financial and monetary system has been in reform, there are many uncertainties in money demand. 

Especially, China’s money demand has its own particularity. This paper studies the determinants of China’s money 

demand through building a linear econometric model and SVAR model. The empirical results show that China’s 

money demand is mainly decided by income, interest rate and expected inflation rate. However, other factors, such 

as financial innovation, government debt, capital mobility and currency substitution, play a relatively small role, 

mainly because China’s financial and monetary system has been under reform. The regression results of sample data 

from different periods show that money demand in China is unstable, indicating that China’s macro-economy has 

certain risks. This finding suggests that China should adopt prudent financial and monetary policies to cope with 

the uncertainty of money demand in the future. 

Ben-Salha and Jaidi (2014) estimated the money demand function in Tunisia. The study assumed that unlike 

many previous money demand studies, the major components of real income are considered. Based on annual data 

ranging between 1979 and 2011 and the ARDL bounds testing approach, results reveal evidence of cointegration 

between the broad money demand and its determinants, namely the final consumption expenditure, the expenditure 

on investment goods, the export expenditure and the interest rate. The error correction model shows that the 

demand for money is only affected by the interest rate and the expenditure on investment goods in the short-run, 

while in the long-run the final consumption expenditure and the interest rate represent the major money demand 

determinants. These findings are robust to a variety of alternative money demand specifications and estimation 

methods. The Saikkonen–Lütkepohl cointegration test with structural shift and the Johansen–Mosconi–Nielsen 

structural break cointegration test are performed in order to control for structural change. In addition, the stability 

of the relationship is checked using the Chow stability test and the Hansen parameter instability test. In the light of 

the study, we advance that monetary policy in Tunisia should be based on a broad definition of money. 

Furthermore, the estimation of money demand functions must take into account the different expenditure 

components of real income. 

Bhattarai (2014) observed that cash in advance and money in utility function models are used to examine 

whether the nature of fluctuations in economic activities and welfare in three interdependent economies are related 

to the stocks and growth rate of money. When the money is exogenously introduced in the form of cash in advance, 

it serves as a medium of exchange and the rate of return in real and nominal assets become equal. Idiosyncratic 

technological shocks generate fluctuations in the growth rates of capital, output, prices, money, consumption, 

investment, labour supply and lifetime utilities of households. When households have money endogenously in their 

utility functions, the stock of money in excess of that required for transactions causes inflation and reduces the 

amount of capital stock and output in these economies. Both CIA and MIU models support for a steady growth rate 

of money according to the growth rate of output. While the inflation targeting by manipulating the interest rates 

for macroeconomic stability is theoretically a prudent policy move, it is impossible for a central bank to eliminate 

business cycles that arise from shocks to production technology or to other structural features of an economy. 

https://www.sciencedirect.com/topics/economics-econometrics-and-finance/cointegration
https://www.sciencedirect.com/topics/economics-econometrics-and-finance/capital-goods
https://www.sciencedirect.com/topics/economics-econometrics-and-finance/monetary-policy
https://www.sciencedirect.com/topics/economics-econometrics-and-finance/macroeconomics


Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
5 

© 2020 Conscientia Beam. All Rights Reserved. 

Iyoboyi and Pedro (2013) studied the narrow money demand function of Nigeria using data from 1970 to 2010, 

using (a) autoregressive distributed lag bounds test approach to cointegration; (b) Augmented Dickey Fuller (ADF); 

(c) Philips–Perron (pp) unit root tests. The result established that cointegration exists among narrow money 

demand, real income, short term interest rate, real expected exchange rate, expected inflation rate, and foreign real 

interest rate in the period under investigation; and that real income is a significant determinant of narrow money in 

both long and short run in Nigeria and interest rate is only significant in the long run nut not short run. Central 

Bank of Nigeria (CBN) (2018) re-examined the stability of the broad money (M2) demand function in Nigeria. The 

study employed Autoregressive Distributed Lag  (ARDL)  as well as bounds  tests in relation to a set of quarterly 

time-series data from 1985(1) to 2016(4). It was then concluded that there was a stable long-run relationship 

between broad money and GDP, stock prices, foreign interest rates and real exchange rate. Darrat (1986) in his 

study of the demand for money function in Nigeria for 1963–1979. The study employed a more involved complex 

form of distributed lag framework for his model specification for currency, narrow money and broad money. 

Consideration was also given to the international monetary influences on domestic money holdings, through the 

inclusion of foreign interest rate, and also to income and expected inflation rate. The result found that income and 

inflationary expectation play significant roles in determining real balances; foreign interest rate exerted a 

significant negative impact on real money demand; and the demand for money function exhibited stability in the 

period under review. Currently, the Central Banks of Nigeria, in the formulation monetary policy, has invariably 

adopted the log-linear real demand for money function (for both broad and narrow money) alongside the partial 

adjustment framework, with the principal independent variables being real income, inflation rate and lagged real 

balances. 

 

3. METHODOLOGY 

3.1. Sources of Data 

Given that both the broad (M2) and the narrow money (M1) were studied. Therefore, quarterly data were 

collected from the Central Bank of Nigeria (CBN) Database and Statistical Bulletin publications as well as from the 

database of the Nigeria Bureau of Statistics (NBS) for all the variables from 2006(Q1) to 2018(Q4). However, 

quarterly data for GDP were not available between 2006 and 2009 though annual series existed, and the annual 

series were subjected to the process of data segregation to obtain estimated quarterly series for GDP. Furthermore, 

variables (M1, GDP and exchange rate) were translated into log form and expressed accordingly 

 

3.2. Estimation and Evaluation Techniques vis a vis Data Desegregation 

Prior to 2010, quarterly data (both at nominal and real prices) were not available for Nigeria’s GDP, except 

annual GDP series. That is, only annual data for those periods (2006 to 2009) were available and collected. 

Incidentally, quarterly and annual data for aggregate credits to private sector (CPS) were available. Therefore, upon 

further analysis it was found that there was a high positive correlation (98%) between annual GDP and credit to 

private sector (CPS). Given the high correlation and the fact that quarterly data on CPS were available for the 

period 2006 to 2009, the annual GDP data were linearly segregated into quarterly data for the period 2006 to 2009 

using quarterly trends of CPS. The linear relationship used for the desegregation is: 

 

Where; qGDP   GDP for a given quarter;  

qCPS   CPS movement for a given quarter;  



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
6 

© 2020 Conscientia Beam. All Rights Reserved. 

Total Movement in CPS= 4th qtr. CPS minus 1st qtr. CPS  

 

3.3. Model Specification 

This study assumes that the demand for money in Nigeria is a function of real GDP, inflation rate, interbank 

lending rate, monetary policy Rate (MPR), Federal Government Treasury Bill rate, rate on savings and exchange 

rate. It then used a linear model of the form in Equation 12: 

1 ( , , , , , , )M f RGDP INFR EXR TBY SDR IBR MPR     (12) 

Where 

M1=Money stock represented by M1. 

RGDP=Real GDP. 

INFR=Inflation rate. 

EXR=Exchange rate.  

TBY=Treasury bill yield. 

SDR=savings deposit rate.  

IBR=Interbank. 

MPR=Monetary Policy Rate. 

The model is expanded in the form (shown in Equation 13): 

0 1 2 3 4 5 6 71 tM RGDP INFR EXR TBY SDR IBR MPR                   (13) 

Where  is the error term. 

Note that M1, RGDP, and TBY are in initially naira form and are converted to their logarithmic forms. Using a 

log model, functional form money demand Equation 14 now becomes: 

0 1 2 3 4 5 6 7ln 1 ln ln tM RGDP INFR EXR TBY SDR IBR MPR                  (14) 

Note that Real GDP (Y) is obtained after deflating by the consumer price index (CPI) 

The OLS model has proved to be a reliable estimation technique for many types of linear models. It is 

particularly suited for many economic variables because economic variables are inherently linear especially over a 

short range. In the class of estimators, OLS estimates are the best linear unbiased linear estimators. That is, the 

estimates of the population parameters computed from linear least squares regression are the most optimal 

estimates; and, it is relatively easy to use. Expectedly, given the different theoretical and empirical works on the 

subject of demand for money, it is expected that the coefficient of income (GDP) will be positive, while coefficients 

of savings and treasury bill yield will be negative. For inflation and exchange rate, their coefficient could either take 

a negative or a positive form depending on the dominant behaviour of economic units. For interbank rate, the 

tendency is for it to be negatively correlated with demand for money.  

Further, it should be noted that Narrow money (M1) is the stock of currency in circulation (C) plus demand 

deposits (D). That is, M1=C+D. This is distinct from broad money given as is the stock of narrow money (M1) as 

well as time, savings and foreign currency deposits with banks. It is given as 

 

Real GDP is a measure of a country’s economic output deflated by the country’s consumer price level (CPI) 

index.  In the money demand function, real GDP is a proxy for wealth. The Exchange Rate is the rate of conversion 

of naira cash assets to dollar denominated cash assets. That is, the naira-to-dollar exchange rate. It serves as a 

proxy for an alternate form of cash assets to naira cash assets. Inflation measures the general level of increase in the 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
7 

© 2020 Conscientia Beam. All Rights Reserved. 

prices of goods and services in an economy over a given period and it proxies the reduction in money value. Interest 

rate: This is cost of borrowing money, or conversely, the income earned from lending money. It is can also refer to 

the return or yield on bond or opportunity cost of deferring current consumption into the future.  It represents the 

speculative motive and it’s proxied, in the model, by interest rate on deposit, Monetary Policy Rate, treasury bill 

yield and interbank lending rate. Monetary Policy Rate (MPR) is the anchor rate set by the Central Bank of Nigeria 

as a reference for other rates while the Treasury bill yield: The yield on treasury bill which equates to the return 

earned on investing in treasury bill. 

 

4. EMPIRICAL RESULTS AND INTERPRETATION 

4.1. Graphical Representation of Money Stock (M1) and Other Variables 

Figure 1 below depicts graphical relationship among the demand for money represented by (M) and the 

exogenous variables: Real GDP(Y), exchange rate (x), inflation (f) and treasury bill (g), savings deposit rate, 

interbank rate and MPR (p). A visual assessment of the trends as shown in the Figure 1 revealed that some of the 

variables (like MPR and interbank rate) do not share almost same appearance as the trend of the dependent variable 

(M1). There is an apparent close movement between M1 and Real GDP. 

 

 
Figure-1. Graphical illustrations of M1 and other variables. 

                              
 

Table-1. Augmented dickey- fuller test results. 

Variable Order of Integration T-ADF Prob* 1% CV 5% CV 10% CV 

M1 I(1) -4.113684 0.0000 -3.584743 -2.928142 -2602225 
RGDP I(0) -5.615693 0.0000 -3.565430 -2.919952 -2.597905 
EXR I(0) -4.685323 0.0000 -3.565430 -2.919952 -2.597905 
INF I(1) -10.77550 0.0000 -3.568308 -2.921175 -2.598551 

INTR I(0) -7.529088 0.0000 -3.565430 -2.919952 -2.597905 
MPR I(1) -6.253837 0.0000 -3.568308 -2.921175 -2.598551 
SDR I(1) -6.345020 0.0000 -3.568308 -2.921175 -2.598551 
TBY I(1) -5.898709 0.0000 -3.568308 -2.921175 -2.598551 

Phillip-Perron Test Results 

M1 I(1) -7.670905 0.0000 -3.568308 -2.921175 -2.598551 
RGDP I(0) -5.615693 0.0000 -3.565430 -2.919952 -2.597905 
EXR I(0) 81965.80 0.0000 -3.565430 -2.919952 -2.597905 
INF I(1) -14.66489 0.0000 -3.568308 -2.921175 -2.598551 

INTR I(0) -7.525145 0.0000 -3.565430 -2.919952 -2.597905 
MPR I(1) -6.307009 0.0000 -3.568308 -2.921175 -2.598551 
SDR I(1) -6.306554 0.0000 -3.565430 -2.919952 -2.597905 
TBY I(1) -6.866263 0.0000 -3.568308 -2.921175 -2.598551 

 

 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
8 

© 2020 Conscientia Beam. All Rights Reserved. 

In Table 1 above, the series were tested for stationarity of individual series using Augmented Dickey Fuller 

(ADF) test, this becomes necessary given the assumption that most economic variables are not stationary at level 

and following the basic requirement of interacting economic variables. Out of the eight variables tested, three 

variables (RGDP, EXR, and INTR) were integrated of order zero I(0), while the remaining five were integrated of 

order one (i.e. I(1)). In all, the study established that each of the series intended for estimating the behaviour of 

demand for money exhibits short run stability. Furthermore, the study attempted to affirm the result of the ADF 

test by using a non-parametric Phillip-Perron test on the same set of the variables, the results showed the same 

three variable with order I(0) and the same five variables with order I(1). Thus, it validates the finding of the ADF 

test.  

Consequent upon the results of the unit root tests that affirmed the existence of short run stability of individual 

variable, it becomes necessary to test for group interaction and affirm a long run relationship. The result in Table 2 

indicates the outcome of Johansen Cointegration (using Trace and Maximum Eigen) test. The result Unrestricted 

Rank (Trace) test indicates that there exist at most three (3) cointegrating equations; these were affirmed by the 

higher values of the Trace statistics when compared with the Critical Values at 5 percent significant level, given 

these, the study therefore rejects the null hypothesis. The result of the Maximum Eigen also led to the rejection of 

the null hypothesis as it indicates that there exist at most two cointegrating equation. We therefore affirm that 

there is a long run relationship among the variable. 

 
Table-2. Johansen cointegration test. 

Series: M1 EXR INF INTR RGDP MPR SDR TBY 

Unrestricted Cointegration Rank Test (Trace)  

Hypothesized  Trace 0.05  

No. of CE(s) Eigenvalue Statistic Critical Value Prob.** 

None * 0.710548 228.6089 159.5297 0.0000 
At most 1* 0.648414 166.6207 125.6154 0.0000 
At most 2* 0.573444 114.3556 95.75366 0.0000 
At most 3* 0.477503 71.75504 69.81889 0.0000 
At most 4 0.327056 39.29828 47.85613 0.0003 
At most 5 0.243387 19.49363 29.79707 0.0156 
At most 6 0.090817 5.548464 15.49471 0.0968 
At most 7 0.015637 0.788043 3.841466 0.2067 

 Trace test indicates 1 cointegrating eqn(s) at the 0.05 level 

 * denotes rejection of the hypothesis at the 0.05 level 

Unrestricted Cointegration Rank Test (Maximum Eigenvalue) 
Hypothesized  Max-Eigen 0.05  

No. of CE(s) Eigenvalue Statistic Critical Value Prob.** 

None* 0.710548 61.98822 52.36261 0.0000 
At most 1 * 0.648414 52.26508 46.23142 0.0000 
At most 2* 0.573444 42.60055 40.07757 0.0011 
At most 3 0.477503 32.45676 33.87687 0.0002 
At most 4 0.327056 19.80465 27.58434 0.0103 
At most 5 0.243387 13.94517 21.13162 0.0853 
At most 6 0.090817 4.760421 14.26460 0.2178 
At most 7 0.015637 0.788043 3.841466 0.1529 

 Max-eigenvalue test indicates no cointegration at the 0.05 level 
 * denotes rejection of the hypothesis at the 0.05 level 

 

 

 

 

 

 

 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
9 

© 2020 Conscientia Beam. All Rights Reserved. 

Table-3. Correlation matrix. 

Variables M1 RGDP INF EXR TBY SDR IBR MPR 

M1 1 0.3812 0.5379 0.7832 0.3591 0.1351 0.1338 0.3959 
RGDP 0.3812 1 -0.0986 0.1343 0.3220 -0.1490 -0.0749 0.0975 

INF 0.5379 -0.0986 1 0.5579 -0.0291 -0.0045 0.0676 0.0380 
EXR 0.7832 0.1343 0.5579 1 0.3381 0.5195 0.1112 0.5494 
TBY 0.3591 0.3220 -0.0291 0.3381 1 0.0578 0.5250 0.7133 
SDR 0.1351 -0.1490 -0.0045 0.5195 0.0578 1 0.0244 0.4432 
IBR 0.1338 -0.0749 0.0676 0.1112 0.5250 0.0244 1 0.4160 

MPR 0.3959 0.0975 0.0380 0.5494 0.7133 0.4432 0.4160 1 
                

4.2. Interpretation of First OLS Result 

Table 3 shows the result of estimation of linear relationship between M (M1) and all the variables. The p-

values (probability values) of four variables (out of seven) are not significant. The p-values of F, G, N and P are 

39.84%, 70.37%, 44.61% and 75.37%. That is, variables F, G, N and P are not significant in explaining the 

movement in demand for money. In other words, more than 50% of our chosen variables (four out of seven) cannot 

explain changes in M1 or demand for money. Given a R-squared value of 76.4% shows a good fit, however the 

Durbin-Watson statistic is 0.58, signifying the presence of autocorrelation. The presence of autocorrelation is 

further confirmed by the Serial Correlation test contained in Table 3. The probability of Chi-squared is less than 

5%, indicating autocorrelation. 

 
Table-4. 1st OLS estimation of money demand function. 

Dependent Variable: M1 

Variables Coefficient Std. Error t-Statistics Probability 

Constant 2.7602 2.4256 1.1380 0.2613 
RGDP 0.4443 0.1499 2.9634 0.0049 

INF 0.1503 0.1763 0.8527 0.3984 
EXR 1.0296 0.1828 5.6317 0.0000 

TBY -0.0499 0.1304 -0.3828 0.7037 
SDR -0.3855 0.1559 -2.4732 0.0173 
IBR 0.0542 0.0704 0.7689 0.4461 

MPR 0.0803 0.2543 0.3157 0.7537 

Other Parameters 
R-Squared 0.7647 Mean dependent Var. 15.5708 
Adj. R-Squared 0.7273 S.D. dependent Var. 0.5084 
S.E. of regression 0.2655 Akaike info criterion 0.3262 
Sum square resid 3.1015 Schwarz criterion 0.6264 
Log likelihood  -0.4812 Hannan-Quinn criterion 0.4413 
F-Statistic 20.4314 Durbin-Watson statistics 0.5797 

Prob(F-statistic) 0.0000   

                          

4.3. Interpretation of Second OLS Result 

Given the poor statistics generated from the above model (contained in Table 4) in terms of p-values of 

coefficients, R-squared, Durbin-Watson and autocorrelation test, the model is re-specified. The re-specification is 

aimed at elimination autocorrelation while obtaining a good fit. The new specification becomes: 

0 1 1 2 3 4 5 61 1t t t t t tM M RGDP INF EXR TBR SDR                

Where the variables are in their log forms,  

M1=Money stock represented by M1. 

RGDP=Real GDP. 

M1t-1= Lagged value of M1. 

INF=Inflation rate. 

EXR=Exchange rate.  



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
10 

© 2020 Conscientia Beam. All Rights Reserved. 

TBR=Treasury bill yield. 

SDR=Savings deposit rate.  

M is lagged by one level and introduced into the model as an independent variable to prevent the presence of 

autocorrelation. The resulting model is shown in Table 5. The p-values of the explanatory variables are significant 

in explaining the variation in M, except F. R-squared of 96.5% shows a good fit and Durbin-Watson of 2.52 

indicates the absence of autocorrelation. This is further confirmed by serial correlation statistics in Table 6 below. 

Therefore, the money demand function (M) estimated by OLS technique becomes: 

11 0.1586 0.8441 1 0.1009 0.0233 0.1436 0.0380t t t t t tM M RGDP INF EXR SDR         

The new model is free from autocorrelation and is hereby presented below: 

 
Table-5. 2nd OLS estimation of money demand function. 

Dependent Variable: M1 

Variable Coefficient Std. Error t-Statistic Probability 

Constant 0.1586 0.8608 0.1842 0.8547 
M1t-1 0.8441 0.0556 15.1791 0.0000 

RGDP 0.1009 0.0546 1.8591 0.0696 
INF -0.0233 0.0613 -0.3807 0.7052 
EXR 0.1436 0.0795 1.8062 0.0776 
SDR -0.0380 0.0563 -0.6753 0.5030 

Other Parameter 
R-square 0.9645 Mean dependent var. 15.5946 
Adj. R-squared 0.9606 S.D. dependent var. 0.4835 
S.E. of regression 0.0960 Akaike info criterion -1.7384 

Sum square resid 0.4149 Schwarz criterion -1.5112 
Log likelihood 50.3297 Hannan-Quinn criterion -1.6516 
F-statistics 244.5680 Durbin-Watson statistic 2.5232 
Prob(F-statistic) 0.0000   

 

 
Table-6. Post estimation diagnostics. 

Serial Correlation Test 
Breuch-Godfrey Serial Correlation LM Test 
Null hypothesis: No serial correlation at up to 2 lags 

F-statistics 2..4639 Prob. F(2,43) 0.0970 
Obs*R-squared 5.2436 Prob. Chi-Squared(2) 0.0727 

Homoscedasticity Test 
Heteroscedasticity Test: Breuch-Pagan-Godfrey 
Null hypothesis: Homoskedasticity 

F-statistics 4.2620 Prob. F(5,45) 0.0029 
Obs*R-squared 16.3898 Prob. Chi-Squared(5) 0.0058 
Scaled explained SS 25.3824 Prob. Chi-Squared(5) 0.0001 

 

 

Conversely, given the presence of lagged values of the dependent variable as regressors, OLS estimation of the 

model that already present itself as an ARDL model will yield biased coefficient estimates; therefore, ARDL bounds 

testing approach developed by Pesaran, Shin, and Smith (2001) was employed to test for presence of long run 

relationship among the variables is hereby required. This procedure, though relatively new method, has many 

advantages over the classical cointegration tests. The bounds tests suggest that the variables of interest are bound 

together in the long-run when narrow money (M1) is the dependent variable. However, the F-Statistics is higher 

than the critical values for I(0) bound but lower than the critical values for I(1) bound. This put the study at a 

crossroad and as well reduces the possibility of having a long run speed of adjustment see Table 7. 

 

 
 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
11 

© 2020 Conscientia Beam. All Rights Reserved. 

Table-7. Cointegration bound test result. 

Null Hypothesis: No long-run relationships exist 

Test Statistic Value k 

F-statistic 2.774954 7 
Critical Value Bounds   

Significance I0 Bound I1 Bound 
10% 1.92 2.89 
5% 2.17 3.21 

2.5% 2.43 3.51 
1% 2.73 3.9 

 

 

Table-8. Short run coefficients. 

Variable Coefficient Std. Error t-Statistic Prob.* 

M1(-1) 0.520696 0.144168 3.611735 0.0023 
M1(-2) 0.356260 0.181161 1.966545 0.0668 

M1(-3) 0.143230 0.169717 0.843936 0.4111 
M1(-4) -0.353142 0.137008 -2.577523 0.0202 
EXR 3.52E-08 1.28E-08 2.753128 0.0141 

EXR(-1) -0.012819 0.131417 -0.097547 0.9235 
EXR(-2) 0.178571 0.131438 1.358594 0.1931 

INF -1.210651 0.598870 -2.021561 0.0603 
INF(-1) 0.294782 0.523537 0.563059 0.5812 
INF(-2) 1.550495 0.420462 3.687597 0.0020 
INF(-3) -1.445361 0.379516 -3.808434 0.0015 
INF(-4) 0.943221 0.306483 3.077558 0.0072 
INTR 0.057933 0.092853 0.623920 0.5415 

INTR(-1) 0.471100 0.184822 2.548939 0.0214 
INTR(-2) 0.663295 0.203296 3.262708 0.0049 
INTR(-3) 0.295069 0.145149 2.032874 0.0590 

MPR -2.993007 2.657931 -1.126066 0.2767 
MPR(-1) 2.521902 2.968475 0.849561 0.4081 
MPR(-2) -7.539781 2.080061 -3.624789 0.0023 
RGDP -0.050160 0.075746 -0.662211 0.5173 

RGDP(-1) 0.126696 0.083963 1.508959 0.1508 
RGDP(-2) 0.121228 0.072817 1.664819 0.1154 
RGDP(-3) -0.295891 0.062053 -4.768385 0.0002 
RGDP(-4) 0.225231 0.069056 3.261586 0.0049 

SDR -0.959672 4.750367 -0.202021 0.8424 
SDR(-1) 5.942464 4.959648 1.198163 0.2483 

TBY 2.919892 1.440714 2.026699 0.0597 
TBY(-1) -2.475186 1.335124 -1.853900 0.0823 
TBY(-2) 2.528519 1.212742 2.084960 0.0535 
TBY(-3) -0.274169 0.919491 -0.298175 0.7694 
TBY(-4) 1.881673 0.832242 2.260970 0.0380 

C 2.430520 3.622738 0.670907 0.5118 

R-squared 0.994485     Mean dependent var 15.65916 

Adjusted R-squared 0.983801     S.D. dependent var 0.420080 
S.E. of regression 0.053466     Akaike info criterion -2.784809 

Sum squared resid 0.045738     Schwarz criterion -1.537341 

Log likelihood 98.83541     Hannan-Quinn criter. -2.313389 

F-statistic 93.07605     Durbin-Watson stat 2.455294 
Prob(F-statistic) 0.000000  

 
 
 
 
 
 
 

 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
12 

© 2020 Conscientia Beam. All Rights Reserved. 

Table-9. Long run coefficients. 

Variable Coefficient Std. Error t-Statistic Prob. 

EXR 0.497820 0.409780 1.214846 0.2420 
INF 0.397905 0.547759 0.726424 0.4781 

INTR 4.467253 2.489339 1.794554 0.0916 
MPR -24.059928 16.212286 -1.484055 0.1572 

RGDP 0.381746 0.581028 0.657018 0.5205 
SDR 14.965340 16.491210 0.907474 0.3776 
TBY 13.757783 10.071146 1.366059 0.1908 

C 7.299834 11.024655 0.662137 0.5173 
 

 

4.4. Policy Implication 

The findings of this study reveal that movements in the prior (lagged) values together with movements in Real 

GDP, inflation, exchange rate and savings deposit rate can explain to a great extent changes in money demand in 

Nigeria. The direction (sign) of variables like Real GDP, inflation, and savings deposit rate conform to expectations.  

The findings suggest that the government, through the central bank, should employ balanced policies that 

encourage economic units to exhibit more confidence in naira assets as against foreign assets and stem the rate of 

dollarization of the economy.  

 

5. CONCLUSION 

The findings of this study reveals that movements in the prior (lagged) values together with movements in 

Real GDP, inflation, exchange rate and savings deposit rate can explain to a great extent changes in money demand 

in Nigeria. Of particular interest is the positive movement in exchange rate. Exchange rate increases as demand for 

money increases, suggesting that more of money demanded finds its way to the foreign exchange market. Citizens 

may have been replacing naira cash assets with dollar cash assets.  The findings from the study suggest that the 

government, through the central bank, should employ balanced policies that encourage economic units to exhibit 

more confidence in naira assets as against foreign assets and stem the rate of dollarization of the economy. The 

study concludes that demand for narrow money in Nigeria is influenced by past demand as well as level of income, 

inflation, exchange rate and savings rate. Interbank lending rate and MPR do not have influence on the level of 

money demand. 

   

6. RECOMMENDATIONS 

Given the findings in this study, the following recommendations are proffered: The monetary authorities, The 

Central Bank of Nigeria, should employ balanced policies that encourage economic units to exhibit more confidence 

in naira assets as against foreign assets and stem the rate of dollarization of the economy 

The Central Bank of Nigeria should develop come with an efficient policy around the components of narrow 

money (that is, currency in circulation and demand deposit) to curb the level of inflation in the country. 

 

Funding: This study received no specific financial support.    
Competing Interests: The authors declare that they have no competing interests.  
Acknowledgement: Both authors contributed equally to the conception and design of the 
study. 

 

REFERENCES 

Baumol, W. J. (1952). The transactions demand for cash: An inventory theoretic approach. The Quarterly Journal of Economics, 

66(4), 545-556. Available at: https://doi.org/10.2307/1882104. 

Ben-Salha, O., & Jaidi, Z. (2014). Some new evidence on the determinants of money demand in developing countries–A case 

study of Tunisia. The Journal of Economic Asymmetries, 11, 30-45. Available at: 

https://doi.org/10.1016/j.jeca.2014.06.001. 



Financial Risk and Management Reviews, 2020, 6(1): 1-13 

 

 
13 

© 2020 Conscientia Beam. All Rights Reserved. 

Bhattarai, K. (2014). Money and economic growth. The Journal of Economic Asymmetries on Science Direct, 11(June), 8-18. 

Busari, D. (2005). On the stability of money demand function in Nigeria. Financial and Economic Review, CBN. 

Central Bank of Nigeria (CBN). (2018). Banks and other financial institutions act 1991 (As amended in 1997, 1998, 1999 and 

2002. 

Darrat, A. F. (1986). Money, inflation, and causality in the north African countries: An empirical investigation. Journal of 

Macroeconomics, 8(1), 87-103. Available at: https://doi.org/10.1016/0164-0704(86)90035-2. 

Dou, X. (2018). The determinants of money demand in China. Cogent Economics & Finance, 6(1), 1564422. 

Essien, E. A., Onwioduokit, E. A., & Osho, E. T. (1996). Demand for money in a debt-constrained economy: A case study of 

Nigeria. Economic and Financial Review, 34(2), 579-605. 

Friedman, M. (1956). The quantity theory of money – A restatement In M. Friedman. Studies in quantity theory of money: Chicago 

University Press. 

Iyoboyi, M., & Pedro, L. M. (2013). The demand for money in Nigeria: Evidence from bounds testing approach. Business and 

Economics Journal, 2013, BEJ-76. 

Keynes, J. M. (1930). The general theory of employment, interest and money: John Maynard Keynes. 

Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied 

Econometrics, 16(3), 289-326. Available at: https://doi.org/10.1002/jae.616. 

Research Department Central Bank of Nigeria. (2016). Monetary policy education in economic series, 2. Nigeria: Central Bank pf 

Nigeria. 

Tobin, J. (1956). The interest-elasticity of transactions demand for cash. The Review of Economics and Statistics, 38(3), 241-247. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Views and opinions expressed in this article are the views and opinions of the author(s), Financial Risk and Management Reviews shall not be responsible or 
answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. 

 


