Crowding Out of Private Sector in Tanzania: Government Expenditure,Domestic Borrowing,and Lending Rates

Abstract

This study empirically analyzes the impact of government expenditure and domestic borrowing on credit to the private sector in Tanzania by increasing lending rates. Quarterly time series data are collected from 2004 to 2018. Autoregressive distributed lag (ARDL) model estimation with a bound cointegration test is used to establish the short- and long-run relationships, and the results are subjected to diagnostic tests for robustness. The result shows that government expenditure and domestic borrowing crowd out credit to the private sector by increasing the lending rate in the long run. This calls for the Tanzanian government to reduce some of its deficit spending and domestic borrowing, and instead look for another way to increase the tax revenue using loans from external sources to fund its budget deficit. Also, the study recommends that the government should put more effort on improving private sector development by making the country an easy place to do business, which in turn will increase the tax base through corporate tax and income tax from business employees.

Keywords

Crowding out lending rates private sector credit ARDL

Introduction

Tanzania is still faced with a high interest rate in such a way that across the sub-Saharan Africa Tanzania has the highest costs of loans recorded in 2018, at 17.50 per cent, higher than the average of 10.89 per cent in the region. The report shows Tanzania has 17.50 per cent loans, followed by Ghana and Angola recorded at 17 and 16.50 per cent, respectively, in 2018 as shown in Figure 1 (Mbani, 2018). Government expenditure and domestic borrowing may partly be responsible for this situation.

To fund the budget in the 2018/2019 financial year, the government borrowed domestically 4.96 trillion shillings in 2017/2018 financial year, which is equal to 80 per cent of the target (annually) of 6.17 trillion shillings. As domestic borrowing tends to increase, the Tanzanian government is paying higher interest rates on treasury bills (T-bills) to attract more auction bids (Dailynews, 2018). This creates panic in the financial market about the government capacity to pay back the loans which leads to the increase in lending rates in commercial banks.

A persistent government overspending and below the target revenue collection weaken the fiscal sustainability, especially in the long term. The adverse fiscal dynamics would be exacerbated by insufficient revenue mobilization and tighter external funding constraints. Domestic financing of a sustained Tanzanian budget deficit caused by external funding tightening and low revenue mobilization may crowd out credit to the private sector (IMF, 2019).

The chairperson of the committee on the 2018/2019 budget Hawa Ghasia said it best that “The committee advise the government to borrow more from the external sources due to the fact that a persisting domestic borrowing will destabilize the economy especially in the private sector since it cannot compete with the government” (Dailynews, 2018).

Therefore, the study aims to find if sustained government expenditure and domestic borrowing has anything to do with persistence increase of lending rates, and if this situation crowds out the private sector participation to the economy by increasing the cost of borrowing.

Figure 1.

Source: Author’s calculation.

Figure 2.

Source: Author’s calculation.

Figure 3.

Source: Author’s calculation.

Literature Review

The impact of government expenditure and domestic borrowing on crowding out the private sector has been addressed by various studies with different methodologies and countries. Assefa (2014) found that foreign liabilities, domestic deposits, real lending interest rate, GDP, inflation, and money supply have a long-run significant impact on banks credit to the private sector in Ethiopia. This abides with another study in Pakistan by Nadeem Aftab (2016) which focused on the impact of interest rates on private sector credit and also found that there is a negative significant impact of interest rates on credit to the private sector. Both studies use the autoregressive distribution lag (ARDL) model for analytical purpose.

This is the same as the study by Amjad Ali (2016) which also took a case study from Pakistan with the data ranging from 1972 to 2015, and using the ARDL technique the study also found that government domestic borrowing negatively affects the amount of credit given to the private sectors. This result abides with that from another study by Sajjad Zaheer (2017) who also analyzed how government borrowing crowds out the private sector in Pakistan using monthly data from 1998 to 2015. The results also suggested that government domestic borrowing significantly affect the private sector negatively.

The study by Molapo (2017) used the ARDL cointegration approach to analyze the factors affecting private sector credit in a short and long run in Lesotho. The study found that the interest rate affects negatively the supply of credit to the private sector in both the short and long run. This result abides with another study by Rahila Munir (2010) which was analyzed using the ARDL cointegration approach to find the short- and long-run impact of various factors affecting the credit to private investment whereby the study also found that interest rates and credit to private sector affect the private investment cementing the crowding out argument. The ARDL method was also used by Gideon Baoko (2017) analyzing the factors influencing private sector credit in Ghana from 1970 to 2011. The study found that real lending rates have a significant impact on credit to the private sector both in the short and long run.

The cross-country model was applied by Mbate (2014) who analyzed the effect of domestic debt on private sector credit across 21 sub-Saharan African countries from 1985 to 2010. Using the generalized method of moments (GMM) system, the study found that domestic debt crowds out the supply of credit to private sectors and capital accumulation.

The structural vector autoregressive technique was used by Phillip Victoria (2017) who studied how an increase in government debt in Nigeria affects credit to the private sector, and found that a persistence increase in government domestic borrowing decreases the amount of credit given to private sector in the economy. The vector autoregressive technique was also used by Shetta and Kamaly (2014) who studied the impact of government borrowing on credit to the private sector and found that, indeed, increase in government domestic borrowing shrinks the amount of credit given to a private sector in Egypt. Impact of government borrowing on crowding out the private sector in South Africa was studied by Kapingura and Biza (2015) who also used the vector autoregressive method on quarterly data and found that in the long run budget deficit negatively affects the private investment. This abides with the previous studies and a priori expectation.

The vector error correction (VECM) model was used by Gerti and Irin (2014) focusing on analyzing the key determinant of bank credit to the private sector in Albania. The study found that lower cost of lending increases bank credit to the private sector. The impact of domestic debt (borrowing) on interest rate was studied by Kariuki (2013) using secondary data from Kenya (Central Bank), and the study concluded that there is a positive relationship between domestic debt and interest rate in Kenya. Also, the study found that the interest rate had a negative significant impact on credit to the private sector. The result abides with another study by Perveen (2017) who found that there is a positive short-run relationship between government domestic debt and interest rate in Pakistan. Granger causality test revealed that government debt and interest rate have unidirectional causality relationship. The results are supported by another study conducted by Gamber (2019), which projects the impact of government debt on interest rate in the USA over the next decade. The study found that the long-run impact of debt on interest rate ranges for 2–3 point for 1 per cent increase in government debt.

The study on the relationship between government expenditure and the interest rate was conducted by Du (2015) who focused on China. The results conclude that there is a positive relationship between government expenditure and interest rate, which abides with the result from this study such that an increase in public expenditure will raise the interest rate. Yi Huang (2017) analyzed the crowding out effect of public debt on private-owned companies in China using the dataset for Chinese local public debt. The results suggested that there is a negative relationship between city-level public debt and private investment. Also, these private companies are more sensitive with the internal cash flow within those cities with high public debt.

Rahman (2016) used the ARDL approach for cointegration and the VECM system for short- and long-run dynamic to analyze the relationship between public investment (government expenditure) and private investment on economic growth in Bangladesh. The study found that public expenditure improves the economy with an increase in private investment, which means public spending that crowd out private sector deteriorates an economy. Omitogun (2018) analyzed the crowding out impact of public expenditure on private sector investment in Nigeria. Using the ARDL approach, the study found that the impact of government expenditure on private sector depends on the components of expenditure such that some types of public expenditure crowd out the private sector, while others crowd in the private sector in Nigeria.

A different result was found by Akpansung (2018) who used the multivariate vector autoregressive method to analyze the impact of domestic debt on credit to the private sector in Nigeria. The study found that there is no significant impact of government domestic debt on credit to the private sector. This result corresponds with another study by Isaya Maana (2008) to a case study in Kenya from 1996 to 2007 and found that there is no significant impact of domestic debt on credit to the private sector, which means no crowding out. Anyanwu (2017) also found that interest rate has no impact on credit to the private sector, but rather it is the government domestic borrowing that has a negative significant crowding out impact on the private sector in 28 oil-dependent countries from 1990 to 2012. The study uses the GMM system and fixed effect estimators.

A study by Demirel (2017) used panel data of Eurozone countries to analyze the impact of government expenditure, borrowing, and interest rate on private investment from 2000 to 2015. The study found that government expenditure, debt, and interest rate all significantly affect private investment negatively and, hence, there is a crowding out effect. A panel cross-country data was also used by Erzen (2008) who studied 85 industrial developing countries from 1980 to 2006 and found that high government debt caused a decline in credit to the private sector in low- and lower-middle-income economies. The impact of government expenditure on private sector using panel data (145 countries from 1960 to 2007) was studied by Furceri and Sousa (2009) who found that government expenditure significantly affects the private sector investment negatively such that there is a crowding out effect. Altin Gjini (2012) focused on eastern European countries from 1991 to 2009 and found that public investment significantly crowds out private sector investment in these countries. Seok-Kyun Hur (2010) focused on Asian countries by analyzing how fiscal policy affects private sector investment. The study found that there is no clear evidence on either crowding out or crowding in effect of fiscal policy on the private sector in these countries.

Mona Esan (2012) in his study on the relationship between government domestic borrowing and private sector credit in Egypt found out that there is more than a one-to-one impact of government borrowing on credit to the private sector such that there are other factors that may cause the private credit to shrink like bank holding on T-bills and securities. These results correspond with those from another study by Al-Majali (2018) who focused on Jordan by analyzing the impact of government domestic borrowing on private sector growth. He also found that there is more than a one-to-one crowding out impact on the private sector such that government domestic borrowing is not the only reason for crowding out the private sector in Jordan. Jordanian bank’s preference to invest its excess liquidity in a low-risk investment may also cause the private sector credit to shrink.

A study in Turkey by Sen and Kaya (2014) found that there is a crowding out effect on the private sector by increasing government transfer spending and interest spending, but there is a crowding in effect when there is an increase in government capital spending in Turkey. Bilgili (2003) also conducted a study in Turkey on the crowding out effect on private sector using the VECM. The study found that fiscal policy had a significant impact on the private sector, both in the short and long run, and concludes that government expenditure crowds out private investment in Turkey. Still, in Turkey, a study by Kustepel (2005) abides with the previous studies on the aspect that government borrowing crowds out private sector growth, but his study also found that government expenditure crowds in the private sector.

Muthu (2017) analyzed whether there is a crowding out or crowding in impact of public investment on private investment in India, using the ARDL approach. The result suggests that government expenditure has a crowding in effect on private sector investment in India such that investment in the public sector had a positive impact on private sector investment.

However, it is crucial to point out that these studies did not focus specifically on crowding out of the private sector in Tanzania. Therefore in a case study of Tanzania, the impact of government expenditure and domestic borrowing on private sector credit through lending rates is empirically analyzed to see if it corresponds to the results on empirical literature reviews.

Methodology

In examining the impact of government expenditure and domestic borrowing on crowding out the credit to the private sector in Tanzania, the study uses a time series quarterly data from 2004 to 2018. The data is collected from secondary sources, whereby credit to the private sector, lending rates, government expenditure, domestic borrowing, and inflation rate data are collected from the Bank of Tanzania quarterly economic review. The equations representing the economic model can be specified as follows:

L e n d i n g = f (l n g e x p e n d, l n i n f l a t i o n)

(1)

L e n d i n g = f (d b o r r, l n i n f l a t i o n)

(2)

C r e d i t = f (l e n d i n g, l n i n f l a t i o n)

(3)

where “lending” represents the lending rates that the banks charge while taking loans, “lngexpend” is the log of government expenditure, “lninflation” is the log of the inflation rate, “dborr” means the domestic borrowing by the government and “credit” means the amount of credit given to the private sector as shown in Table 1.

To test for stationary, the study uses augmented Dickey–Fuller (ADF) test and Phillips Perron (PP) unit root test. Akaike information criterion (AIC) is used for choosing the number of lags to be used in the model. Therefore the generalized ARDL (p, q_¹, q_²) can be specified as follows:

\begin{array}{l} L e n d i n g_{t} = α_{0} + \sum_{i = 1}^{p} β_{1} Δ l n g e x p e n d_{t - 1} \\ + \sum_{j = 0}^{p} β_{2} Δ l n i n f l a t i o n_{t - 1} + ε_{t} \end{array}

(4)

\begin{array}{l} L e n d i n g_{t} = α_{0} + \sum_{i = 1}^{p} β_{1} d b o r r_{t - 1} \\ + \sum_{j = 0}^{p} β_{2} Δ l n i n f l a t i o n_{t - 1} + ε_{t} \end{array}

(5)

\begin{array}{l} C r e d i t_{t} = α_{0} + \sum_{i = 1}^{p} β_{1} l e n d i n g_{t - 1} \\ + \sum_{j = 0}^{p} β_{2} Δ l n i n f l a t i o n_{t - 1} + ε_{t} \end{array}

(6)

where β_¹ and β_² are the coefficients to be estimated; p, q are optimal lag orders, and _^t is the error term. Bound test for cointegration was performed to estimate the long-run relationship between the variables. The hypothesis for the test is that H_⁰: β_¹ = β_² = 0 against the alternative hypothesis H_¹: β_¹ ≠ β_² ≠ 0. F-statistics, t-statistics, and p-value will be a criterion for making a decision such that for the model to have cointegration the F-statistics and t-statistics should be greater than the critical values for upper and lower bounds. p-Value should be statistically significant below the 10 per cent level. If that is confirmed, then the error correction model (ECM) can be specified as follows:

Table 1.

Variable Name, Definition, Source, and a priori Hypothesis

Variable	Definition	Source	A Priori Expectations
Credit	Credit provided to private sector (% growth)	Bank of Tanzania
Lending	Lending rates (% growth)	Bank of Tanzania	—
Lngexpend	Log of government expenditure	Bank of Tanzania	—
Dborr	Government domestic borrowing (% growth)	Bank of Tanzania	—
Lninflation	Log of inflation (% growth)	Bank of Tanzania	—

Source: Author’s own.

\begin{array}{l} L e n d i n g_{t} = α_{0} \\ + \sum_{i = 1}^{p} β_{1} l e n d i n g_{t - 1} \\ + \sum_{i = 1}^{p} β_{2} Δ l n g e x p e n d_{t - 1} \\ + \sum_{j = 0}^{p} β_{3} Δ l n i n f l a t i o n_{t - 1} \\ + λ E C T_{t - 1} + ε_{t} \end{array}

(7)

\begin{array}{l} L e n d i n g_{t} = α_{0} + \sum_{i = 1}^{p} β_{1} l e n d i n g_{t - 1} + \\ \sum_{i = 1}^{p} β_{2} d b o r r_{t - 1} + \sum_{j = 0}^{p} β_{3} Δ l n i n f l a t i o n_{t - 1} \\ + λ E C T_{t - 1} + ε_{t} \end{array}

(8)

where ECT = ( $l e n d i n g_{t - 1} - θ X_{t}$ )

\begin{array}{l} C r e d i t_{t} = α_{0} + \sum_{i = 1}^{p} β_{1} c r e d i t_{t - 1} \\ + \sum_{i = 1}^{p} β_{2} l e n d i n g_{t - 1} \\ + \sum_{j = 0}^{p} β_{3} Δ l n i n f l a t i o n_{t - 1} \\ + λ E C T_{t - 1} + ε_{t} \end{array}

(9)

where ECT= ( ${credit}_{t - 1} - θ X_{t}$ ).

Note that

Λ = is the speed of adjustment with a negative sign

$θ = \frac{\sum_{i = 0}^{q} β_{i}}{α}$ is the long-run parameter

β_¹, β_², and β_³ are the coefficients representing short-run dynamics of the model’s adjustment in the long run.

Estimation Results

Time series modeling involves testing for unit root. That is, variable estimated must be stationary to avoid spurious regression. Therefore Table 2 shows the unit root tests (ADF and PP) both nonstationary in levels and first difference. The results suggest that lending rates, credit to private sectors, and domestic borrowing are stationary in level while government expenditure and inflation are stationary in first difference.

Table 2.

Unit Root Test

Variable	H0: Nonstationary in Levels		H0: Nonstationary in First Difference
Variable	ADF Statistics	PP Statistics	ADF Statistics	PP Statistics
Lending	−3.614***	−3.526***
Lngexpend	−1.783	−1.382	−19.303***	−24.957***
Lninflation	−1.297	−1.596	−6.648***	−6.676***
Credit	−4.951***	−4.956***
Dborr	−5.539***	−5.550***

Source: Author’s own.

Notes: ***, **, and * Denote significance at 1%, 5%, and 10%, level, respectively, for the ADF and PP test of unit root. The null hypothesis is that the series are nonstationary.

Table 3.

Autoregressive Distributed Lag (ARDL) Approach: Dependent Variable is Lending Rate

Model	ARDL(2,1,2)	ARDL(1,2,0)
Variables	Lending	Lending
L.lending	0.494***	0.540***
	(0.149)	(0.117)
L2.lending	0.312*
	(0.157)
Dlngexpend	1.373**
	(0.635)
L.dlngexpend	0.911
	(0.634)
Dlninflation	−1.064	−0.946
	(0.705)	(0.623)
L.dlninflation	1.585**	1.212**
	(0.681)	(0.582)
L2.dlninflation	−1.036*	−1.416**
	(0.602)	(0.546)
Dborr		0.000726*
		(0.000363)
Constant	3.059	7.099***
	(1.876)	(1.805)
No. of observations	49	53
R-squared	0.607	0.562

Source: Author’s own.

Notes: Standard errors in parentheses.

***p < 0.01, **p < 0.05, and *p < 0.1.

Table 3 shows the ARDL models with stationary variables representing the impact of government expenditure and domestic borrowing to lending rates in Tanzania. Starting with the first column the results suggest that government expenditure has a positive and statistically significant impact on lending rates. The positive coefficient implies that when other factors remain constant if the government expenditure increases by 1 per cent then lending rates will increase by 1.373 per cent. Shifting to the second column the results suggest that domestic borrowing has also a positive and significant impact on lending rates in Tanzania, and its coefficient implies that when other factors remain constant if domestic borrowing increases by 1 per cent then lending rates will also increase by 0.0726 per cent, which eventually decreases credit to the private sector. These results abide with a priori hypothesis and other studies by Rahman (2017), Anthony Anyanwu (2017), and Karanja (2013).

Although the government expenditure and inflation have unit root problem in the short run, there might be a long-run relationship between government expenditure, inflation, and lending rates in Tanzania. Thus, a bound test for cointegration is performed in Table 4. The results suggest that there is cointegration among the variable due to the fact that t-statistics and F-statistics are above the 10 and 5 per cent critical value, respectively, and the p-value is lower than 10 per cent. This is also a case in Table 5 where the bound cointegration test is performed for the long-run relationship between lending rates and domestic borrowing.

Table 4.

Bound Test for Cointegration: Relationship Between Lending Rate and Government Expenditure

Bound test for cointegration
Critical value bounds of the F-statistics and p-value
10%		5%		1%		p-Value
I(0)	I(1)	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)
3.231	4.270	3.936	5.092	5.556	6.951	0.021	0.064
−2.550	−3.210	−2.876	−3.564	−3.522	−4.254	0.005	0.031
F-statistics = 4.811t-Statistics = −3.779Sample size = 58K is the number of regresses = 2

Source: Author’s own.

Table 5.

Bound Test for Cointegration: Relationship Between Lending Rate and Domestic Borrowing

Bound test for cointegration
Critical value bounds of the F-statistic and p-value
10%		5%		1%		p-Value
I(0)	I(1)	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)
3.254	4.262	3.959	5.075	5.572	6.903	0.007	0.022
−2.564	−3.225	−2.886	−3.575	−3.525	−4.255	0.001	0.011
F-statistics = 5.993t-Statistics = −4.228Sample size = 58K is the number of regresses = 2

Source: Author’s own.

Table 6 shows the results from the ECM, and they suggest that government expenditure and domestic borrowing have a positive and significant impact on the lending rates in Tanzania. The positive coefficient implies that when other factors remain constant if government expenditure and domestic borrowing increase by 1 per cent then lending rate will increase by 0.968 per cent and 0.137 per cent, respectively. The results correspond with the a priori hypothesis of the study, and another study by Karanja (2013) also found a positive relationship between government domestic borrowing and interest rate in Kenya.

Table 6.

ARDL with Error Correction Model (ECM) Approach: Dependent Variable is Lending Rate

Model	ARDL(2,1,2)	ARDL(2,1,0)
Variables	Lending	Lending
ECM(−1)	−0.603***(0.159)	−0.519***(0.127)
Short run
LD.lending	−0.102
	(0.135)
D.lngexpend	−0.007
	(0.416)
D.lninflation	−0.151	−0.921
	(0.542)	(0.576)
LD.lninflation	1.188**	1.122*
	(0.548)	(0.573)
Long run
Lngexpend	0.968***
	(0.229)
Lninflation	−1.004***	−0.941**
	(0.328)	(0.454)
Dborr		0.00137**
		(0.000626)
Constant	5.858***	8.943***
	(2.043)	(2.181)
Diagnostic test
Hettest (p-value)	0.5248	0.3448
BGodfrey (p-value)	0.7888	0.3148
DW-statistics	2.02	2.12
Ramsey test (p-value)	0.2792	0.4342
Multicollinearity (VIF)	1.66	1.26
No. of observations	53	55
F-statistics	4.59	4.37
R-squared	0.391	0.372

Source: Author’s own.

Notes: Standard errors in parentheses.

***p < 0.01, **p < 0.05, and *p < 0.1.

The error correction coefficients ECM(−1) in both columns are negative and highly significant as expected because it shows the long-run relationship between lending rates and government expenditure in the first column and between the lending rate and domestic borrowing in the second column. And the coefficients in absolute terms shows that speed of adjustment of government expenditure and domestic borrowing will re-establish in the long run. In our case, it is 0.60 and 0.52 per cent for the first and second column, respectively.

Diagnostic test justifies the efficiency, robustness, and accuracy of the result; therefore, the model is subjected to following tests: Ramsey RESET specification test, heteroskedasticity test (Hettest), multicollinearity test (vif), and autocorrelation test (DW-statistics and BGodfrey). The results suggest that there is no heteroskedasticity, serial correlation, and specification problem.

CUSUM square stability test has been applied to test for stability among coefficients. The graphical representation of stability check is shown in Figures 2 and 3, whereby the plots are found inside the critical bounds suggesting that there is stability on the ECMs among coefficients and with that the null hypothesis is rejected.

In this section, the study now analyzes the impact that the lending rates have on credit to private sectors. From Table 7, which represents the ARDL (1,1,0) model, we see that lending has no significant impact on credit to the private sector in Tanzania. This calls for bound cointegration test as represented in Table 8, and from the results we see that there is cointegration (long-run relationship) between credit to private sector and lending rates in Tanzania due to the fact that t-statistics, F-statistics are above the 10 and 5 per cent critical value, and p-value is lower than 1 per cent. Therefore the ECM is an appropriate model for estimation.

Table 7.

ARDL (1, 1, 0) Approach Representing the Impact of Lending Rate to Private Credit

Variables	Coefficient	Standard Error	p-Value
L.credit	0.275*	(0.139)	0.049
Lending	−0.930	(0.661)	0.149
L.lending	−0.994	(0.667)	0.139
Dlninflation	−1.903	(3.248)	0.508
Constant	33.90***	(10.30)	0.002
No. of observations	51
R-squared	0.321

Source: Author’s own.

Notes: Standard errors in parentheses.

***p < 0.01, **p < 0.05, and *p < 0.1

Table 9 represents the ECM and the results suggest that in the long run, the lending rate has a negative and significant impact on credit to the private sector, which means that in Tanzania higher lending rates reduces the amount of credit given to the private sectors. The coefficient implies that when other factors are fixed, then a 1 per cent increase in lending rates decreases the credit to the private sector in Tanzania by 2.398 per cent in the long run. The results abide with a priori hypothesis of the study and other studies by Assefa (2014), Nadeem Aftab (2016), Gideon Baoko (2017), and Molapo (2017).

Table 8.

Bound Test for Cointegration: Relationship Between Private Credit and Lending Rate

Bound test for cointegration
Critical value bounds of the F-statistic and p-value
10%		5%		1%		p-Value
I(0)	I(1)	I(0)	I(1)	I(0)	I(1)	I(0)	I(1)
3.263	4.255	3.966	5.061	5.570	6.865	0.000	0.000
−2.571	−3.233	−2.891	−3.580	−3.524	−4.253	0.000	0.000
F-statistics = 12.203t-Statistics = −6.027Sample size = 58K is the number of regresses = 2

Source: Author’s own.

Table 9.

ARDL with Error Correction Model (ECM) Approach: Dependent Variable Is Credit to Private Sector

Variables	Coefficient	Standard Error	p-Value
ECM(−1)	−0.763***	(0.126)	0.000
Long run
Lending	−2.398***	(0.698)	0.001
Lninflation	−1.655	(1.519)	0.285
Short run
D.lending	0.985	(0.611)	0.107
Constant	34.87***	(0.611)	0.002
Diagnostic test
Hettest (p-value)	0.7052	Multicollinearity (VIF)	1.46
BGodfrey (p-value)	0.9131	F-statistics	9.20
DW-statistics	1.98
Ramsey test (p-value)	0.4537
No. of observations	55
R-squared	0.404

Source: Author’s own.

Notes: Standard errors in parentheses.

***p < 0.01, **p < 0.05, and *p < 0.1.

The error correction coefficient is negative and highly significant as expected because it shows the long-run relationship between credit to the private sector and lending rates in Tanzania, and the coefficients in absolute terms shows that speed of adjustment of lending rates will re-establish in the long run by 0.763 per cent. The diagnostic test implies that there is no heteroskedasticity, serial correlation, and specification problem.

In conclusion, the government expenditure and domestic borrowing in Tanzania crowd out credit to the private sector in the long run by increasing lending rates. Therefore, the result of this study recommends that the Tanzanian government should consider reducing its domestic borrowing to fund its budget deficit, and instead look for another way to increase the tax revenue and loans from external sources. Also, the study recommends that the government should put more effort on improving private sector development by making the country an easy place to do business, which in turn will increase the tax base through corporate tax and income tax from business employees.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

References

Akpansung

A. O.

(2018). Analysis of the impacts of domestic debts on private sector credit, lending rate, and real output: Evidence from Nigeria. Journal of Finance and Economics , 6(3), 111–123.

Al-Majali

A. A.

(2018). Crowding out effect of public borrowing: The case of Jordan. International Review of Management and Market , 8(1), 119–125.

Anthony Anyanwu

C. G.

(2017). Government domestic debt, private sector credit, and crowding out effect in oil-dependent countries. Journal of Economic Research , 22, 127–151.

Assefa

(2014). Determinants of growth in bank credit to the private sector in Ethiopia: A supply side approach. Research Journal of Finance and Accounting , 5(17), 90–102.

Baki Demirel

C. E.

(2017). The crowding out effect from the European debt crisis perspective: Eurozone experience. International Journal of Sustainable Economy , 9(1), 1–18.

Bilgili

(2003). Dynamic implications of fiscal policy: Crowding-out or crowding-in? (MPRA Paper 24111). Munich, Germany: University Library of Munich.

Dailynews . (2018, June 19). Tanzania: Extensive domestic borrowing is harmful . Retrieved from Tanzania Daily News: https://allafrica.com/stories

(2015). Positive correlation between government expenditure and real interest rate: Testing Ramsey Model based on American and Chinese data. Procedia Economics and Finance , 30, 202–212.

Erzen

(2008). The determinants of financial development and private . Ankara, Turkey: Middle East Technical University.

10.

Furceri

Sousa

D. F.

(2009). The impact of government spending on the private sector: Crowding-out versus crowding-in effects (NIPE Working Paper No. 6). Braga, Portugal: University of Minho

11.

Gamber

(2019). The effect of government debt on interest rate (Working Paper 2019-01).

12.

Gerti

Irin

(2014). Determinants of bank credit to private sector: The case of Albania. International Journal of Economic Policy in Emerging Economies , 8(1).

13.

Kapingura

Biza

R. A.

(2015). Do budget deficits crowd out private investment? An analysis of the South African economy. International Journal of Economic Policy in Emerging Economies , 8(1), 52–76.

14.

Gideon Baoko

I. A.

(2017). Determinants of bank credit in Ghana: A bounds-testing cointegration approach. African Review of Economics and Finance , 9(1), 33–61.

15.

IMF. (2019). United Republic of Tanzania: Staff reportfor the 2019 article IV consultation . New York, NY: International Monetary Fund.

16.

Isaya Maana

R. O.

(2008). Domestic debt and its impact on the economy—The case of Kenya . In Annual African Econometric Society Conference, Petroria, South Africa.

17.

Kariuki

K. G.

(2013). Relationship between domestic debt and interest rate in Kenya . Nairobi, Kenya: University of Nairobi.

18.

Karanja

K. G.

(2013). Relationship between domestic debt and interest rate in Kenya . Nairobi, Kenya: University of Nairobi.

19.

Kustepel

(2005). Effectiveness of fiscal spending: Crowding out and/or crowding in? YÖNET M VE EKONOM , 12(1), 185–192.

20.

Matiur Rahman

A. M.

(2016). Public versus private investment in economic growth of Bangladesh: An econometric analysis. International Journal of Trade and Global Markets , 9(3), 228–248.

21.

Mbani

(2018, December 17). Tanzania borrowers ‘pay the most’ . Retrieved from The Citizen: www.thecitizen.co.tz

22.

Mbate

(2014). Domestic debt, private sector credit and economic growth in sub-saharan africa. African Development Review , 25(4), 434–446.

23.

Molapo

(2017). An econometric approach to private sector credit in Lesotho. The Journal of Business in Developing Nations , 15(1), 32–54.

24.

Muthu

(2017). Does public investment crowd-out private investment in India. Journal of Financial Economic Policy , 9(1), 50–69.

25.

Nadeem Aftab

K. J.

(2016). Impact of interest rate on private sector credit; Evidence from Pakistan. Jinnah Business Review , 4(1), 47–52.

26.

Omitogun

(2018). Investigating the crowding out effect of government expenditure on private. Journal of Competitiveness , 10(4), 136–150.

27.

Perveen

(2017). Impact of total, internal and external government debt on interest rate in Pakistan (MPRA Paper 83427). Munich, Germany: University Library of Munich.

28.

Phillip

(2017). The impact of domestic debt on private credits in Nigeria: A structural VAR approach (1981–2015). Journal of Social Sciences , 3(1), 64–72.

29.

Rahila Munir

R. U.

(2010). Investment, savings, interest rate and bank credit to the private sector nexus in Pakistan. International Journal of Marketing Studies , 2(1), 140–146.

30.

Rahman

A. A.

(2017). Impact of government borrowing on financial development (A case study of Pakistan) (Munich Personal RePEc Archive No. 78100). Munich, Germany: University Library of Munich.

31.

Sajjad Zaheer

F. K.

(2017). Does government borrowing crowd out private sector credit in Pakistan (SBP Working Paper Series No. 83). Karachi, Pakistan: State Bank of Pakistan, Research Department.

32.

Sen

Kaya

H. Ş

. (2014). Crowding-out or crowding-in? analyzing the effects of government spending on private investment in Turkey. Panoeconomicus , 61(6), 631–651.

33.

Seok-Kyun Hur

S. M.

(2010). Fiscal policy and crowding out in developing Asia (Asian Development Bank Economics Working Paper No. 222).

34.

Shetta

Kamaly

(2014). Does the budget deficit crowding-out private credit from the banking sector. The Case of Egypt. Topics in Middle Eastern and African Economies , 16(2), 2014

35.

Yi Huang

M. P.

(2017). Local crowding out in China (EIEF Working Papers Series 1707). Rome, Italy: Einaudi Institute for Economics and Finance (EIEF).