Index

1. INTRODUCTION

In the economics literature demand for real balances is also referred to as demand for money. Due to its importance in macroeconomic management, demand for real balances has a great historical value amongst economists and policy makers. In Nigeria,  the debate about the significance of  demand function for real balances and its determining factors can be traced back to Tomori (1972); Ajayi (1974); Teriba (1974); Ojo (1974) and Odama (1974) also known as ‘TATOO’ debate.  Like in other countries, the debate has continued to strive in Nigeria with researchers contributing to the existing knowledge about improving the conduct of monetary policy and its transmission as well as its proper stabilization functions.

Nigeria provides a motivating case study of the Central Bank of Nigeria (CBN), termed as a money targeter, which implies that for monetary targeting to be efficacious, the stability of real balances demand must be met (Tule et al., 2018). Therefore, in spite of having a better and robust understanding of the demand for real balances, it was more germane to CBN’s objective to achieve operational excellence. . Moreover, considering the demand for real balances as a dynamic phenomenon due to the  variation in the determinants of money demand (Tule et al., 2018)  which is due to the  transformational forces in structural, institutional, and regulatory domains, scholars have sought to re-examine the parameter stability of long-run function for real balances demand. This is in view of such forces like output expansion, financialization, development, financial innovation, financial globalization, and uncertainty that characterize a modern economy.  Nonetheless, the effect of government spending on real balances demand has not been completely elucidated despite rich empirical works on money demand function and its stability in developed and developing economies. Previous empirical studies relating to government sector has focused more on the nexus between budget deficit and real balances demand (Yellen, 1989; Aamir, 2015; Ibrahim, 2018). Increasing government spending associated with modern economy, and supported by Keynesian demand management ideas, however, causes a change in aggregate demand along with fiscal deterioration in the form of  high budget deficits and public debt problem (Bank for International Settlements, 2012). Additionally, government spending is procyclical in developing countries (Culha, 2017) as a result of which monetary authorities face a challenging task of managing destabilizing effects of government spending (Ebadi, 2018) . Thus, keeping in view that asymmetric government spending relates to cyclical behavior of fiscal policy or changes in government expenditure, this paper aims to investigate the effects of asymmetric government spending on real balances demand in Nigeria for monetary policy purposes.

Theoretically, economists view income or real GDP, representing an economic activity, as a scale variable in the real balances demand function (Laidler, 1993). Using Barro (1990) spending model, real GDP is decomposed into the government and private sectors to properly measure the influence of government spending on real balances demand (Ebadi, 2018) . In accordance with what is generally done in literature involving variable selection and the framework choice to estimate the demand for real balances, this paper has introduced two scale variables in the model namely  public spending and capital stock., This is combined with financial innovation and opportunity costs determinants of holding real balances and includes  interest rate, inflation, and exchange rate (Groessl and Tarassow, 2015; Berentsen et al., 2017; El-Rasheed et al., 2017; Ebadi, 2018; Gerlach and Kugler, 2018; Kayongo and Guloba, 2018; Tule et al., 2018). In light of the intuition from empirics, this paper will apply the Autoregressive Distributed Lag (ARDL) cointegration technique to estimate the long-run relationship between variables in the model. While the stability of demand for real balances function in Nigeria will be discussed using the CUSUM and CUSUMSQ tests.

The rest of the paper is structured as follows: Section two presents a review of literature on real balances demand. Section three discusses the model and the estimation strategy while the empirical results are presented in Section four. Finally, section five summarizes the conclusion of the study.

2. REVIEW OF LITERATURE

The central goal of growing literature on demand for money is to extend the Keynesian, Neo-Keynesian and Neoclassical analysis of money demand dynamics in the light of observable developments in an economy and methodological innovations. To this end, Folarin and Asongu (2017) have investigated the stability of Nigeria’s money demand function in the aftermath of financial sector deregulation using CUSUM and CUSUMSQ tests. The results confirm stable demand for money during the period of 1992Q1 to 2015Q4, while their findings from ARDL co-integration test support the views of  Bahmani-Oskooee and Bahmani (2015) about inflation rate being a more suitable measure of an opportunity risk for less developed financial environment. Aliha et al. (2018) have studied the influence of financial innovation on the United States’ money demand. Adopting GARCH approach, the paper finds that financial innovation, specified as external shock, does not have impact on volatility of money demand in the US during 1990 to 2016.

Séraphin (2018) considered the relationship between financial market development and liberalization on money demand in Central African Economic and Monetary Community (CEMAC) economies ¹ using a generalized method of moments (GMM) framework for panel data of six-member countries over the period 1981 to 2015. The study found negative relationship between financial liberalization and money demand. Bassey et al. (2017) using error correction modeling technique as well as the Chow (1960) test of stability, revealed that increase in return to other money assets such as Savings deposit, Equity and Treasury bill reduces economic agent’s desire to hold money. In part, the stability test result shows stability of money demand for Nigeria during 1986 to 2013. Gerlach and Kugler (2018) assess the Swiss money demand under free banking during the period before the institutionalization of the Swiss National Bank. They found the banking activity as an important determinant of the long-run money demand during the reviewed period.

Akinlo and Emmanuel (2017) examined the role of stocks in the Nigeria’s money function using co-integration approach with structural break and tests for the stability of the function. The findings confirm that stock prices have a significant positive influence on long run narrow and broad money demand whilst the test suggests absence of stability in the demand for money function from 1986Q1 to 2012Q4. Based on the findings, the study affirms the argument by Baharumshah et al. (2009) that if the effects of asset prices (stock) are ignored, the stabilizing policy of monetary authorities can cause uncertainty in the domestic economy. Tule et al. (2018) re-assessed money demand and its stability in Nigeria using the Autoregressive Distributed Lag (ARDL) bounds testing approach. They found a significant and positive effect of stock prices on the long-run broad money demand, which according to the authors is suggestive of financialization and integration of domestic economy into the global economic system.  The results also confirm a stable money demand function during the period 1985Q1 to 2016Q4.

Bahmani-Oskooee and Maki-Nayeri (2018) suggest an asymmetric relationship between policy uncertainty and money demand. Using Korea, Australia and U.S. data, respectively, they demonstrated that people relatively hold more cash during periods of decreased uncertainty as compared to times of increased uncertainty, indicating buffer activity against an uncertain future. Aworinde (2018) explores the non-linearity and uncertainties in the stability for demand for money in Nigeria during 1960-2015 with the inclusion of output uncertainty and monetary uncertainty measures, consistent with the findings of  Bahmani-Oskooee and Baek (2016). This study adopts a nonlinear autoregressive distributed lag (NARDL) approach and the results of the analysis show that unlike monetary volatility, output volatility have no significant effect on demand for money in Nigeria both in short-run and long run. Also, the tests establish stability of the demand for money function. Using Nigeria’s data covering 1980 to 2014, El-Rasheed et al. (2017) examined the influence of monetary uncertainty on the stability of money demand function. The results from the ARDL bound testing shows co-integration between broad money and monetary uncertainty as well as a unidirectional causality running from former to later.

Employing a micro-founded model of money demand under uncertainty within an intertemporal optimizing risk-averse households’ context, Groessl and Tarassow (2015) showed that increased demand for money of households can be explained by positive changes in inflation risk and capital market risk based on U.S. data covering 1986M1 to 2007M12. Specifically, this study found a positive reaction of households' money demand to a positive capital market risk change and a higher inflation risk. This demonstrates how uncertainty transmits to macroeconomic developments. Kayongo and Guloba (2018) tested the influence of generalized autoregressive conditional based economic uncertainty on the stability of money demand function in Uganda during 2001Q4 to 2017Q2, a period described as financial liberalization. Using ARDL, the study discovered money demand stability and found that economic uncertainty does not impact M1 in the short-run but negatively affects all money demand balances including base money, , indicating effect of portfolio diversification in M1 and M2 after one quarter..

Nyumuah (2018) examines interest and exchange rates volatilities making an impact on money demand in developing economies using data of Equatorial Guinea, Gambia, Nigeria and Uganda. The paper found no significant effects of interest rate and exchange rate volatilities on money demand in these economies. Nwude et al. (2018), narrowing on Nigerian data during the period 1991Q1 to 2014Q4 and employing the ARDL-ECM approach, confirmed a cointegrating association between the real broad money demand, real income, domestic interest rate, inflation rate, exchange rate and foreign interest rate. They also found stable long-run money demand function using the CUSUM and CUSUMSQ test. In another study, based on Masters and Rodríguez-Reyes (2005) framework of precautionary demand for money, Rodríguez-Reyes (2018) developed a search model that explains the companies and criminal clients trade relationship in order to investigate how the interaction between the firm’s demand for private security and crime indirectly affects the demand for money. The study found that companies’ demand for private security makes an impact on the aggregate demand for money, but remotely and further demonstrated the indirect channel between crime and money arising from market externality.

Berentsen et al. (2016) strengthened the money demand in a micro-founded monetary model by introducing limited commitment to repay loan factor. They posit that depending on monetary policy, limited commitment will create endogenous borrowing constraint thereby influencing the shape of the money demand curve. Thus, using model calibration approach and the money demand data for the UK, Australia and Canada, they revealed that limited commitment significantly enhances the robustness of the money demand function. Tweneboah and Alagidede (2018) employed a long-run money demand function within the portfolio balance framework, by distinguishing between currency substitution and capital mobility as well as taking account for the pre and post inflation targeting monetary policy regime to investigate the effect of currency substitution on demand for money in Ghana. Based on the results from ARDL-ECM approach, the study reported empirical evidence suggesting no significant influence of currency substitution, exchange rates and foreign interest rates on the domestic money demand dynamics during the period 1960 to 2013.

Taofik (2017) turned a searchlight on short and long-run implications of budget deficit on aggregate money demand or the money market equilibrium in general using cointegration analysis and ECM approach. The results established a positive and significant relationship between money demand and budget deficit in the short- and long-term horizon for the period 1980 to 2015. The findings of the study indicated the validity of the Keynesian and Neoclassical propositions for Nigeria. Ebadi (2018) offered a complementary view of the government spending effect on money demand. The study suggested income disaggregation in the public and the private sectors, which was hinged on Barro (1990) spending model. Using U.S. data and ARDL approach, the results established a positive significant effect of government spending on money demand.

How valid is this supposition in Nigeria? Therefore, the main aim of this paper is to investigate the effect of government spending on the demand for money in Nigeria. However, unlike the Ebadi (2018) ARDL  specification, the current study introduced asymmetric analysis of government spending to the model and also considered the impact of  financial innovation and expected inflation instead of  actual CPI in the system.  It will be motivating and valuable to gain some insight on the money demand channel of government spending measure in Nigeria.

3. METHODOLOGY

Model specification:

Standard demand for money function comprises interest rate and inflation to denote opportunity costs. The inclusion of nominal exchange rate account for currency substitution (Bahmani-Oskooee et al., 2015) of which an estimate of 𝛾 in Equation 1 could be either positive or negative. Financial innovation measure is used to capture the financial sector liberalization and fintech advances while national income or gross domestic product (GDP) is a scale variable representing activity in the real economy and the model specification, as shown in Equation 1:

Where lnRM is the logarithm of real M2 (broad money supply), ln is the logarithm of GDP, NTB is the 91-day Nigeria Treasury bill rate, Infex is the expected inflation rate, BDCExr is the logarithm of Bureau de Change exchange rate, FinInno is the financial innovation ratio, c is the constant term, and ε is the error term.

Borrowing Ebadi (2018) specification, the study decomposes national income in public sector and private sector and rearranges the demand for money in Equation 2 shown as thus:    

Where lnRGe is the logarithm of government spending and lnRKS is the logarithm of capital stock. Since government spending could have positive or negative effect on the demand for real balance, an estimate of 𝛼 could be negative or positive.

To extend the above symmetric analysis to asymmetric analysis, this paper followed (Bahmani-Oskooee and Maki-Nayeri, 2018) and decomposed the lnRGE variable into two new time-series variables shown in Equation 3 as thus:

Where POSt, which is the partial sum of positive changes, reflects only growth in government spending, and NEGt indicates the partial sum of negative changes in government spending which captures only reductions in government spending. Therefore, by replacing lnRGE in ARDL model (2) by POS and NEG variables, we get the following specifications shown as in Equation 4:

4. EMPIRICAL ANALYSIS

Using Phillip-Perron unit root test at its constant, constant and trend and without constant and trend form in testing the stationarity of variables in the model, the result is presented in Table 4.1:

Table-4.1. Philip-Perron (PP) Unit Root Test Result.

Notes: (1) D denotes the first difference operator; (2) *** indicates significance at the 1% level.

Using PP test, Table 4.1 reveals that INFex is stationary at level with constant, constant and trend and without constant and trend; while FinInno is also stationary at level with and without constant and trend. Hence, it is sufficient to conclude that INFex and FinInno are integrated in order zero I(0). After taking the first difference, all other variables LM, LGE, LKS, NTB and BDCExr are stationary at first difference and are therefore said to be integrated in order one I(1).

Table-4.2. Bounds test for Linear Cointegration.

Source: Extract from Results.

Table 4.2 reports the results of cointegration / F-statistic at the 5% level of significance. This confirms the presence of a long run relationship between real balances demand as the dependent variable in the model and explanatory variables for the period under consideration in Nigeria. This is because the calculated F statistic at 6.34 is greater than upper and lower bound  values  of 2.17 and 3.21, respectively, at 5% significance level (Pesaran et al., 2001) and thus, infers that there exists a co-integrating relationship among the time series in the level form, without considering whether they are I(0) or I(1). In other words, the Null hypothesis of no cointegration can be rejected at 5% significance level because F test statistic is greater than critical upper bounds value I(1). We can proceed for the ARDL model estimation.

Table-4.3. Estimated long-run coefficients of real balances demand model from asymmetric government spending effect.

Notes: ***, ** and * denote statistical significance at the 1%, 5% and 10% levels, respectively.

Table 4.3 presents the ARDL estimation results of the response of demand for money to positive and negative government expenditure, capital stock, treasury bills, inflation expectation, exchange rate, and financial innovation. The result reveal that, one period of real broad money supply or the dependent variable [LRM (-1)] shows a positive and significant impact on itself, while the positive value of real government spending (LRGE_POS) also shows a positive and significant impact. However, the negative value of real government spending (LRGE_NEG) depicts a negative and significant impact on the dependent variable. Effect of real capital stock (LRKS) is seen as mixed; it is positive and insignificant in its current value, but negative and significant in its one and two period lags. Treasury bills and inflation expectation have a positive and negative influence, respectively, but both are insignificant, while exchange rate shows a negative and significant impact. Financial innovation depicts a negative and significant impact in its current and various lags except lag one, which is positive and insignificant.

We now turn our attention to discuss the results of short-run coefficient estimates of the model. The result presented in Table 4.4 suggests that in the short run, the positive value of real government spending is positive and significant, while its negative value shows a positive but insignificant impact on real money supply. Exchange rate remains negative and insignificant, while financial innovation also depicts negative and significant impact on the dependent variable. The effect of   NTB and INFEX is instantaneous; however, the dynamic responses will be controlled by dependent variable. Hence, the short run parameter is equivalent to the instantaneous parameter in the long run ARDL result.  The coefficient of the error correction term suggests 14 per cent speed of adjustment to any disequilibrium in the short run. In other words, the estimated ECT (-1) is equal to 0.14 which states that the departure from the equilibrium is adjusted by 14%. It is negative, significant and less than one, which means that information from this, can be relied upon for policy decisions.

Table-4.4. Estimated short-run coefficients of real balances demand model from asymmetric government spending effect.

Notes: ***, ** and * denote statistical significance at the 1%, 5% and 10% levels, respectively.

However, the reliability of the model depends on the performance of the diagnostic tests as presented in Table 4.5.

Table 4.5. Diagnostic Statistics.

Source: Extract from Results.

The outcome of the diagnostic tests such Breusch-Godfrey Serial Correlation LM Test, Ramsey’s RESET test, Normality Test and Heteroscedasticity test reveal that, the model passed all diagnostic tests except the normality test as there was no evidence of serial correlation and heteroscedasticity. While the RESET test implies the correctly specified ARDL model and the Jarque-Bera based on normality test shows that the residuals are not normally distributed. The stability of the regression coefficients is tested using the cumulative sum (CUSUM) of the recursive residual test for structural stability. Plots of the CUSUM suggest that the regression equation are stable considering that the CUSUM test statistic does not go beyond the bounds of the 5% level of significance as presented in Figure 4.1 and Figure 4.2.

5. CONCLUSION

In this paper, we have contributed to the debate on the demand for real balances function due to  changes in government spending. In the context of cyclical behavior of fiscal policy, and particularly pro-cyclical government spending in developing countries (Culha, 2017) the Barro (1990) spending model enabled this  study to decompose real GDP into government and private sectors and properly measure the influence of asymmetric government spending on real balances demand (Ebadi, 2018) . The result, using ARDL model for Nigeria, revealed that the positive value of real government spending has a positive and significant impact while the negative value of real government spending makes a negative and significant impact on the long-run real money supply. The findings suggest that positive and negative value of real government spending accounts for about 30 per cent while  long-run changes in real money account for 20 percent . The other determinants of demand, particularly real balances in the long-run, exchange rate showed a negative and significant impact. The negative effect of exchange rate met the theoretical expectation that as domestic currency depreciates the demand for domestic currency also declines.  Similarly, financial innovation depicted a negative and significant impact in its current and various lags except lag one, which is positive and insignificant. This supports the disposition that the growth in financial innovation induces a shift to less liquid assets from more liquid ones thus lowering the demand for real balances. Treasury bills rate and inflation expectation have a positive and negative influence on real balances demand, respectively, but both are insignificant. Both positive effect of Treasury bill rate and negative effect of inflation expectation (validating monetary policy short-term orientation) are in line with the view of Fisher’s equation which states that increasing interest rates requires an increase in the rate of money growth and the responsiveness of inflationary expectations to monetary condition. The evidence on degree and direction of asymmetry offers a fresh dimension to policymakers to the need for efficient policy coordination to maximize the benefits of stabilization policies.

REFERENCES

Aamir, S., 2015. The relationship and impact of money growth and budget deficit on inflation in Pakistan. VFAST Transactions on Education and Social Sciences, 5(2): 29-36.

Ajayi, S.I., 1974. The demand for money in the Nigerian economy: Comments and extensions. Nigerian Journal of Economic and Social Studies, 16(1): 165-173.

Akinlo, A.E. and M. Emmanuel, 2017. Stock prices and demand for money: Evidence from Nigeria. Review of Business & Finance Studies, 8(1): 1-19.

Aliha, P.M., T. Sarmidi and F.F. Said, 2018. Investigating the impact of financial innovation on the volatility of the demand for money in the United Stated in the oontext of an arch/garch model. Regional Science Inquiry, 10(1): 19-26.

Aworinde, O.B., 2018. Output uncertainty, monetary uncertainty and the Nigerian demand for money. Acta Universitatis Danubius. Economica, 14(3): 173-185.

Baharumshah, A.Z., S.H. Mohd and M.A. Yol, 2009. Stock prices and the demand for money in China: New evidence. Journal of International Financial Markets, Institutions and Money, 19(1): 171-187.Available at: https://doi.org/10.1016/j.intfin.2007.11.001.

Bahmani-Oskooee, M. and J. Baek, 2016. Economic uncertainty, monetary uncertainty and the Korean demand for money. Journal of Economic Policy Reform, 20(1): 86-97.Available at: 10.1080/17487870.2016.1213166.

Bahmani-Oskooee, M. and S. Bahmani, 2015. Nonlinear ARDL approach and the demand for money in Iran. Economics Bulletin, Access Econ, 35(1): 381-391.

Bahmani-Oskooee, M., S. Bahmani, A. Kones and A.M. Kutan, 2015. Policy uncertainty and the demand for money in the United Kingdom. Applied Economics, Taylor & Francis Journals, 47(11): 1151-1157.Available at: https://doi.org/10.1080/00036846.2014.993138.

Bahmani-Oskooee, M. and M. Maki-Nayeri, 2018. Policy uncertainty and the demand for money in Korea: An asymmetry analysis. International Economic Journal, 32(2): 219-234.Available at: https://doi.org/10.1080/10168737.2018.1481129.

Bank for International Settlements, 2012. Fiscal policy, public debt and monetary policy in emerging market economies. No 67 in BIS Papers. Available from https://www.bis.org/publ/bppdf/bispap67.pdf .

Barro, R.J., 1990. Government spending in a simple model of endogeneous growth. Journal of Political Economy, 98(5, Part 2): S103-S125.Available at: https://doi.org/10.1086/261726.

Bassey, E.N., S.U. Udo and O.U. Etim, 2017. Modelling the demand for money function in Nigeria: Is there Stability?. Bulletin of Business and Economics (BBE), 6(1): 45-57.

Berentsen, A., S. Huber and A. Marchesiani, 2016. Limited commitment and the demand for money, ECON - Working Papers No. 199, Department of Economics - University of Zurich.

Berentsen, A., S. Huber and A. Marchesiani, 2017. Limited commitment and the demand for money. The Economic Journal, 128(610): 1128-1156.Available at: https://doi.org/10.1111/ecoj.12449.

Chow, G.C., 1960. Tests of equality between sets of coefficients in two linear regressions. Econometrica, 28(3): 591-605.Available at: https://doi.org/10.2307/1910133.

Culha, A.A., 2017. Asymmetric government expenditure: A comparison of advanced and developing countries. Journal of Economic Policy Reform: 1-20.Available at: https://doi.org/10.1080/17487870.2017.1374863.

Ebadi, E., 2018. On the effect of government spending on money demand in the United States: An ARDL cointegration approach, MPRA Paper No. 86399.

El-Rasheed, S., H. Abdullah and J. Dahalan, 2017. Monetary uncertainty and demand for money stability in Nigeria: An autoregressive distributed lag approach. International Journal of Economics and Financial Issues, 7(1): 601-607.

Folarin, O. and S. Asongu, 2017. Financial liberalization and long-run stability of money demand in Nigeria. MPRA Paper No. 81190. Available from https://mpra.ub.uni-muenchen.de/81190/.

Gerlach, S. and P. Kugler, 2018. Money demand under free banking: Switzerland 1851-1906. Swiss Journal of Economics and Statistics, 154(1): 18.Available at: https://doi.org/10.1186/s41937-017-0013-8.

Groessl, I. and A. Tarassow, 2015. A microfounded model of money demand under uncertainty, and some empirical evidence. Macroeconomics and Finance Series 201504, University of Hamburg, Department of Socioeconomics, revised Jan 2018.

Ibrahim, T., 2018. Budget deficit-money demand nexus in Nigeria: A myth or reality. MPRA Paper No. 86265.

Kayongo, A. and A. Guloba, 2018. Economic uncertainty and money demand stability in Uganda during financial Liberalization: A garch and ARDL approach. Applied Economics and Finance, 5(4): 70-86.Available at: https://doi.org/10.11114/aef.v5i4.3271.

Laidler, D., 1993. The demand for money: Theories, evidence and problems. New York: Harpercollins College Publishers.

Masters, A. and L.R. Rodríguez-Reyes, 2005. Endogenous credit-card acceptance in a model of precautionary demand for money. Oxford Economic Papers, 57(1): 157-168.Available at: https://doi.org/10.1093/oep/gpi009.

Nwude, C., O. Offor and S. Udeh, 2018. Determinants and stability of money demand in Nigeria. International Journal of Economics and Financial Issues, Econ Journals, 8(3): 340-353.

Nyumuah, F.S., 2018. The impacts of interest rate and exchange rate volatilities on the demand for money in developing economies. International Journal of Economics and Finance, 10(3): 56-69.Available at: https://doi.org/10.5539/ijef.v10n3p56.

Odama, J.S., 1974. The demand for money in the Nigerian economy: Some comments. Nigerian Journal of Economic and Social Studies, 16(1): 178-188.

Ojo, O., 1974. The demand for money in the Nigerian economy: Some comments. Nigerian Journal of Economic and Social Studies, 16(1): 149-152.

Pesaran, M.H., Y. Shin and R.J. Smith, 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.

Rodríguez-Reyes, L.R., 2018. A model of the indirect effect of crime on the demand for money. Mexican Journal of Economics and Finance New Era REMEF, 13(4): 571-584.Available at: https://doi.org/10.21919/remef.v13i4.339.

Séraphin, P.Y., 2018. Financial liberalization and money demand in CEMAC: Evidence from GMM estimations of a dynamic panel. International Journal of Economics and Financial Research, 4(8): 250-257.

Taofik, I., 2017. Budget deficit-money demand nexus in Nigeria: A myth or reality? MPRA Paper No. 86265, University Library of Munich, Germany.

Teriba, O., 1974. The demand for money in the Nigerian economy: Some methodological issues and further evidence. Nigeria Journal of Economic and Social Studies, 16(1): 153-164.

Tomori, J., 1972. The demand for money in the Nigeria. Nigerian Journal of Economic and Social Studies, 16(1): 153-164.

Tule, M.K., U.M. Okpanachi, P. Ogiji and N. Usman, 2018. A reassessment of money demand in Nigeria. CBN Journal of Applied Statistics, 9(1): 47-75.

Tweneboah, G. and P. Alagidede, 2018. Currency substitution and stability of money demand in Ghana. The Journal of Developing Areas, 52(2): 41-53.

Yellen, J.L., 1989. Symposium on the budget deficit. Journal of Economic Perspectives, 3(2): 17-21.

Footnote: