|
Inflation Anchoring, Exchange Rate and Sectoral Economic Growth in Nigeria
|
David Mautin Oke1, Augustus Onyokwonu2
Abstract: The Nigerian economy has been faced with high inflation rate, accompanied with a devastated exchange rate and a slow growth rate of the economy. The authors, in this paper, have examined inflation anchoring, exchange rate and sectoral economic growth nexuses in Nigeria. A multivariate cointegration approach was adopted. The result shows that inflation anchoring with a rising exchange rate would have contractionary effects on economic growth, agricultural output, industrial output and trade output in Nigeria. However, such policy would not have significant impacts on the building and construction sectors as well as service sector of the economy. The inflation threshold for Nigeria has been found to be 9 percent. Thus, inflation anchoring policy may be meaningful in Nigeria if exchange rate is well managed and the sub sectors of the economy are developed. A piecemeal disbursement of loan facilities for accountability, at low interest rate to the agricultural, industrial and trade sectors remains promising in promoting the growth of the sectors, while a low inflation rate is pursued and secured. Studies in future on the subject may essentially examine how inflation anchoring would affect economic growth in Nigeria without interacting inflation anchoring with exchange rate in their models.
Keywords: Inflation anchoring; Exchange rate; Sectoral economic growth; Cointegration
JEL Classifications: E31; F31; F43; C30
1. Introduction
The Nigerian economy has always been faced with the challenge of high level of inflation. This has made its central bank to continue to work hard to reverse the trend over the years. In the past four years, the central bank has strived to reduce the country’s inflation rate from 15.68 percent in 2016 to 12.56 percent as at June 2020. The high inflation rate in Nigeria is accompanied with a devastated exchange rate and a slow growth rate of the economy. In order to critically manage an economy in this 21st century, it has been established that central banks should not reduce inflation while allowing the economic growth to suffer. Interestingly, achieving low inflation would produce economic growth in the long-run. It is also not a gainsaying that the Nigerian economic performance has suffered from sharp practices and loopholes that have been created in the system, among other factors which have negatively affected the exchange rate. A high inflation rate makes interest rate to be expensive thereby slowing down economic growth. Inflation anchoring is one policy window for achieving low inflation. Where a low inflation is pursued while other macroeconomic variables are allowed to fluctuate, inflation is said to be anchored.
In adopting the framework of inflation anchoring, responsibilities of the players must be spelt out in order to assess their transparency and accountability. As well known, in an open economy, a fluctuating exchange rate in a country affects its inflation rate. This is known as the exchange rate pass-through impact. The size of this impact affects central bank as it determines whether it should concentrate on controlling nominal exchange rate depreciation pressure which threatens price stability. Some studies such as Hattori & Yetman (2017) and Gondo & Yetman, (2018) have shown that the features of the pass-through might affect the rate of inflation which a central bank target.
Many empirics have shown that low inflation spurs economic growth (e.g. Fischer, 1993; Barro, 1995). The authors of these works, using both cross-sectional and panel data have found that high levels of inflation are injurious to economic growth either based on national or regional front. In recent times, it is becoming difficult for authors to arrive at this conclusion because inflation has become lower than it was in the 1970s and 80s in most countries of the world, yet economic growth has not increased as expected. Therefore, the conclusion on the inflation-economic growth nexus is becoming fragile. As a result, studies must continue to interrogate the debate. For example, the study by Gondo & Yetman (2018) concludes that there are no robust connections between economic growth and inflation. Similarly, a study by Hattori & Yetman (2017) concludes that evidences on the gains of low inflation to the economy are not very convincing.
Studies have continued to employ aggregate data to assess the causal impact of inflation on economic growth while including relevant control variables. This paper seeks to overcome the limitation of aggregate variables by using sectoral contributions to output and applying a difference-in-difference strategy such as the one used in the study by Cecchetti & Hakkio (2009). This has hardly been done for a developing country like Nigeria. The study differs from other studies on Nigerian economy by looking at how anchoring of inflation in the midst of volatile exchange rate can stimulate sectoral contributions to outputs
Following this section, part two reviews the relevant literature while in section three, the methodology is discussed. Section four presents the findings and discuss it while in the fifth part, conclusion and recommendations are drawn.
2. Literature Review
Inflation anchoring is based on the new Keynesian monetary policy drive. The policy framework is majorly characterized by official announcement of the targeted inflation rate for a stated time period; monetary policy design that aims at low and stable inflation rate as well as perceived accountability and transparency of the central bank (De Pooter, Rabitaille, Walker & Zdinak, 2014). This anchoring is flexible and uses discretionary monetary policy instrument to achieve low inflation. It is expected that central bank will act quickly to bring down inflation rate to avoid much variability in inflation compared to output. However, an economic shock such as a fall in oil price will raise inflation rate. The Keynesian has shown that there is no permanent trade-off between inflation and economic growth. According to the school, factors such as fiscal and monetary policy factors, expectation, labour wage and other prices will promote inflation and output growth in the short-run. Though, in the long-run, these factors and shock on the steady state of the economy lead to dynamic adjustment such that inflation-output variability trade-off will be positive initially but after some time will turn negative (Dornbusch, Fischer & Kearney, 1996). The monetarists concentrate more on the long-run dynamics of supply-side of the economy as against the short-run side (Dornbusch et al., 1996). They argued that increase in money supply or its velocity in excess of output growth is what causes inflation in the long-run.
Solow (1956) and Swan (1956) first advocated the neoclassical discourse on the subject. The conclusion of the school is missed. For example, Tobin (1965) concludes that inflation and output growth nexus is positive but Stockman (1981) argues otherwise. The endogenous growth model also supports a negative relationship of inflation among other variables with economic growth since inflation reduces investment return thereby reducing capital formation which reduces growth. Meanwhile, the depreciation of the exchange rate will increase economic growth (Marshall, 1923; Lerner, 1944; Speller, 2006). This Marshall-Lerner condition will happen where the addition of price elasticity of export demand and that of import demand is larger than unity. The Keynesian framework has shown that devaluation will enhance output through its influence on aggregate demand. Both IS-LM developed by Hicks (1937), advanced by Hansen (1953) and the Mundell–Fleming IS-LM developed by Mundell (1963) and Fleming (1962) support the exchange rate depreciation as having a positive effect on economic growth. However, exchange rate may not exhibit a complete pass-through (Obstfeld & Rogoff, 1995).
A number of studies on inflation anchoring employed survey data in their analysis. This method captures history on the subject and is independent of risk of secondary error. However, survey data analysis is not strong on the measurement of the extent of causal relationship between expected inflation and the actual inflation. Kozicki & Tinsley (2012) combined short-run and long-run expected inflation from their survey and actual inflation in an internal consistent manner to estimate inflation dynamic of the United States. Many studies did not enforce such internal consistency but employed econometric methods to compute inflation forecasts across different prospects. For example, Mehrotra & Yetman (2014) used a Weibull decay model. In this approach, the model-implied bound of the expected inflation is construed as the targeted inflation. Using the same method, Gondo & Yetman (2018) found anchored inflation to lie within the target inflation for all the countries examined.
Demertzis & Viegi (2009) find well-anchored inflation expectations in the euro zone at levels in consonance with the Euro system’s target. This result is similar to the study by Cruijsen & Demertzis (2011) on the same eurozone while adopting a vector autoregressive approach. However, the study by Strohsal & Winkelmann (2015), using exponential smooth transition autoregressive dynamics method for European Monetary Union (EMU), US, Sweden and UK reveal a considerable varying degree of inflation anchoring across the region/countries and expectation levels.
The study by Audu & Amaegberi (2013) has found interest rate to have positive effect on inflation in Nigeria while increase in exchange rate shows a negative impact on the economy. Meanwhile, Ncube and Ndou (2011) using a sign restrained method of Bayesian VAR shows an inverse response of real rate of interest to inflation shock. The study further supports the Fisher effect and the argument for firm inflation targeting to promote economic growth. Besides, inflation targeting mechanism that is flexible and attaches relevance to real effective exchange rate also enhances economic growth. As noted earlier, inflation targeting is an important component of inflation anchoring. However, the conditions for targeting inflation must be met especially by emerging economies for them to achieve the targets set and by extension anchor inflation well. The foregoing review of the literature suggests a further need for analysis on how inflation anchoring and exchange rate would impact sectoral growth in Nigeria.
3. Theoretical Framework and Methodology
Our methodology draws from the theoretical background of rational inflation expectation theory which was propounded by John F. Muth in 1961 (Muth, 1961). The theory was later popularized in macroeconomics by Robert Lucas and Thomas Sargent. According to the rational expectation hypothesis, individuals use present obtainable and pertinent information in establishing their expectations. They refuse to rely on any past experience (Shaw, 1983). The theory says that expectations are rational since they efficiently include all available present information when the expectation is framed and not just the previous information.
In our inflation anchoring model, we access the degree of inflation anchoring by following the works of Levin, Natalucci & Piger (2004) and Choi, Furceri & Loungani (2018). Our equations are as follows:
(1)
Where = the first difference of inflation expectations in h years, that is, h future period. The baseline year employed is the medium-term (5 years ahead) while a robustness check is conducted by expanding the scope of the period into short term and long-term horizons (1 and 10 years ahead respectively). Reverse causality and multicollinearity are minimized because expected inflation in medium-term are less collinear with lagged and current inflation levels.
The term measures inflation shocks. By definition, it is the difference between inflation expectations in short-term and actual inflation from the CBN statistical bulletin. The coefficient measures the extent of anchoring in h- years-ahead inflation expectations. It is called “shock anchoring” (Ball & Mazumder, 2011). The smaller the value of the coefficient, the well anchored inflation expectation is while the reverse is the other case. Therefore, after estimating the shock, the degree of inflation anchoring is the coefficient .
In addressing the interactive effect of the impact of the degree of inflation anchoring and exchange rate on various sectors’ output performance, our procedures are as follows:
For the agricultural sector, we draw from the work of Yaqub (2010) and Mordi, Adebiyi, Adenuga, Adebayo, Abeng, Akpan, Evbuomwan. (2013) and we specify that:
(2)
Where t is the time period and is stochastic disturbance term. Other variables are denoted as follows:
AGRIC = Agricultural contribution to GDP
IA = Inflation anchoring degree
EXCH = Exchange rate
PLR = prime lending rate
CPS = Credit to private sector
GOVCAP = Government capital expenditure
The a priori expectations of the regression coefficients are
For the industrial sector, we draw from the studies by Mordi et al. (2013) and we specify that:
(3)
Where IND = Industrial sector’s contribution to GDP and other variables are as earlier defined
The a priori expectations of the coefficients are
For the building and construction sector, we specify that:
(4)
Where BUILD = Building and Construction sector’s contribution to GDP and INV = Domestic investment while other variables’ definitions remain.
The a priori expectations of the coefficients are
For the trade (Wholesale and Retail) sector, we specify that:
(5)
Where TRADE = Wholesale and Retail trade’s contribution to GDP and other variables are as previously defined under equation (4).
The a priori expectations of the coefficients are
For services sector, we specify that:
(6)
Where SERV = Services sector’s contribution to GDP while other definitions of variables remain
The a priori expectations of the coefficients are
We proceed to determine the threshold at which inflation affects the GDP. We noticed that threshold regression model was advanced by Khan & Senhadji (2001) for the analysis of inflation threshold for developing and developed countries. However, we adopt the model of Bawa & Abdullahi (2011) to estimate inflation threshold that affects Nigeria’s GDP. Our GDP model to account for the inflation threshold effect is stated as follows:
(7)
Dt: 1 for INFt > K
0 for INFt ≤ K
Where INFt = Current year inflation rate, K = Threshold inflation
Data used which cover 1981-2017 were sourced from the Central Bank of Nigeria Statistical Bulletin, 2017.
4. Discussion of Findings
We begin with the discussion of the preliminary results of descriptive statistics in Table 1. It reveals a sharp depreciation of the exchange rate from a minimum of 0.61 to a maximum rate of 305.79 over the studied period. It further shows that the highest credit to private sector was 21.1 trillion naira while the least was 8.57 billion naira. The Jacque Bera normality test shows that given the acceptance rejection criteria, agricultural output contribution to GDP, exchange rate, GDP, government capital expenditure, industrial output contribution to GDP and prime lending rate were normally distributed since their probability values were more than 5% significance level. Nevertheless, the null hypothesis was not accepted for building and construction contributions to GDP, trade contribution to GDP, service contribution to GDP, domestic investment and credit to private sector.
Table 1. Descriptive Statistics
Variables |
Mean |
Median |
Maximum |
Minimum |
Skewness |
Jarque-Bera |
Prob |
Obs |
AGRIC |
7.43 tr |
4.70 tr |
17.2 tr |
2.30 tr |
0.649674 |
4.491192 |
0.105 |
37 |
BUILD |
1.00 tr |
0.679 tr |
26.8 tr |
0.336 tr |
1.241479 |
9.537317 |
0.008 |
37 |
CPS |
4.19 tr |
0.431 tr |
21.1 tr |
8.57bill |
1.369181 |
11.73851 |
0.003 |
37 |
EXCH |
82.78627 |
92.69335 |
305.7901 |
0.610025 |
0.713608 |
3.167107 |
0.205 |
37 |
GDP |
32.7 tr |
22.4 tr |
69.0 tr |
13.8 tr |
0.801593 |
5.099944 |
0.078 |
37 |
GOVCAP |
0.39 tr |
0.270 tr |
1.16 tr |
4.10 bill |
0.625666 |
3.946986 |
0.139 |
37 |
IND |
9.39 tr |
8.56 tr |
13.8 tr |
5.26 tr |
0.136120 |
2.593838 |
0.273 |
37 |
INV |
5.41 tr |
2.87 tr |
16.9 tr |
1.80 tr |
1.253209 |
9.852748 |
0.007 |
37 |
PLR |
17.59486 |
17.58000 |
29.80000 |
7.750000 |
0.189738 |
0.727478 |
0.695 |
37 |
SERV |
10.2 tr |
6.45 tr |
25.4 tr |
3.67 tr |
0.964841 |
6.264668 |
0.044 |
37 |
TRADE |
4.70 tr |
2.63 tr |
11.7 tr |
1.66 tr |
0.969762 |
6.535250 |
0.037 |
37 |
The denotation tr and bill mean trillion and billion in Nigerian naira
Source: Authors’ computation
Next is the unit root test using the Augmented Dickey Fuller test developed by Dickey & Fuller (1981) which result in Table 2 shows that all the variables were stationary at first difference. Based on this result, a cointegration analysis is conducted to ascertain whether an equilibrium relationship exists among the variables. Two approaches were employed for the cointegration which are – the Johansen and the Engle-Granger tests
Table 2. Augmented Dickey-Fuller (ADF) Unit Root Test Result
* Indicates significant at 5%, thus, the variable is stationary at that order
** Indicates significant at 1%, hence, the variable is stationary at that order
Source: Authors’ Computation
The Johansen test is a multivariate generalization of the ADF test which examines the linear cointegration of variables (Johansen, 1991; Maddala & Kim, 1998). It is a maximum likelihood estimation approach that makes it possible to estimate all co-integrating vectors when there are more than two variables. If there are three variables each with unit roots, there are at most two cointegrating vectors. An absence of a long run relationship or equilibrium relationship between the variables coincides with zero coefficients for the variables in the model. A rejection of null hypothesis indicates that there is a long-run relationship.
Table 3. Johansen Co-integration Test Result
Models according to objectives |
Hypothesized No. of Cointegrating Equations |
Eigen Value |
Trace Statistics |
5% Critical Value |
Probability |
Aggregate Output Equation |
None ** |
0.620730 |
79.79092 |
69.81889 |
0.0065 |
At most 1 |
0.523392 |
45.85814 |
47.85613 |
0.0761 |
|
At most 2 |
0.306086 |
19.92098 |
29.79707 |
0.4282 |
|
At most 3 |
0.164646 |
7.131745 |
15.49471 |
0.5624 |
|
At most 4 |
0.023582 |
0.835246 |
3.841466 |
0.3608 |
|
Agricultural Sector Equation |
None * |
0.620210 |
73.19566 |
69.81889 |
0.0262 |
At most 1 |
0.421369 |
39.31087 |
47.85613 |
0.2481 |
|
At most 2 |
0.273683 |
20.16269 |
29.79707 |
0.4119 |
|
At most 3 |
0.223958 |
8.970759 |
15.49471 |
0.3681 |
|
At most 4 |
0.002755 |
0.096568 |
3.841466 |
0.7560 |
|
Industrial Sector Equation |
None * |
0.705030 |
73.19400 |
69.81889 |
0.0262 |
At most 1 |
0.359973 |
30.46308 |
47.85613 |
0.6951 |
|
At most 2 |
0.238894 |
14.84452 |
29.79707 |
0.7901 |
|
At most 3 |
0.135499 |
5.290146 |
15.49471 |
0.7773 |
|
At most 4 |
0.005528 |
0.194029 |
3.841466 |
0.6596 |
|
Building and Construction Equation |
None * |
0.622082 |
73.71602 |
69.81889 |
0.0236 |
At most 1 |
0.459695 |
39.65833 |
47.85613 |
0.2350 |
|
At most 2 |
0.249394 |
18.11156 |
29.79707 |
0.5578 |
|
At most 3 |
0.161830 |
8.070966 |
15.49471 |
0.4579 |
|
At most 4 |
0.052629 |
1.892261 |
3.841466 |
0.1689 |
|
Wholesale and Retail Trade Equation |
None ** |
0.696969 |
85.07232 |
69.81889 |
0.0019 |
At most 1 |
0.480913 |
43.28510 |
47.85613 |
0.1258 |
|
At most 2 |
0.309298 |
20.33615 |
29.79707 |
0.4003 |
|
At most 3 |
0.151029 |
7.384509 |
15.49471 |
0.5334 |
|
At most 4 |
0.046156 |
1.653935 |
3.841466 |
0.1984 |
|
Service Sector Equation |
None * |
0.636283 |
71.71794 |
69.81889 |
0.0350 |
At most 1 |
0.425113 |
36.31968 |
47.85613 |
0.3804 |
|
At most 2 |
0.241770 |
16.94431 |
29.79707 |
0.6442 |
|
At most 3 |
0.170696 |
7.257412 |
15.49471 |
0.5479 |
|
At most 4 |
0.019984 |
0.706532 |
3.841466 |
0.4006 |
* Connotes significant at 5%, hence, the number of co-integrated equations
** Indicates significant at 1%, thus, the numbers of co-integrated equations
Source: Authors’ Computation
Table 3 presents the Johansen co-integration result for the 6 models as specified in the preceding section. The first equation is used to examine how inflation anchoring affects GDP, that is, the aggregate output. The second, third, fourth, fifth and sixth equations examined the effect of inflation anchoring on agriculture, industry, building and construction, wholesale and retail trade as well as services sector contribution to GDP, respectively. The result revealed that at 5% significance level, there is one co-integrating vector among the variables in the aggregate output model and all the 5 disaggregated models. Therefore, there was no reason to accept the null hypothesis of no co-integration among the variables. However, Table 4 reveals that one of the equations is not co-integrated using the Engle Granger co-integration test. This approach was developed in the research of Engle and Granger (1987).
Table 4. Engle Granger Co-integration Test Result
Model |
Dependent |
tau-statistic |
Prob.* |
z-statistic |
Prob.* |
Aggregate Output Equation |
LOG(GDP) |
-8.997238 |
0.0176 |
-34.74673** |
0.0002 |
Agricultural sector equation |
LOG(AGRIC) |
-8.234166 |
0.0054 |
-36.28530* |
0.0122 |
Industrial Sector Equation |
LOG(IND) |
-2.189016 |
0.9087 |
-291.5898** |
0.0000 |
Building and Construction Equation |
LOG(BUILD) |
-11.424890 |
0.0000 |
-45.21998** |
0.0000 |
Wholesale and Retail Trade Equation |
LOG(TRADE) |
-10.212141 |
0.0000 |
-37.07917* |
0.0227 |
Service Sector Equation |
LOG(SERV) |
-2.797728 |
0.7078 |
-11.41291 |
0.8068 |
* Indicates significant at 5%, hence the equation is co-integrated
** Indicates significant at 1%, thus the equation is co-integrated.
Source: Authors’ Computation
In the main section, we estimate inflation anchoring. Following the model specified in equation (1), the estimated result is presented below:
(1.938)
The baseline result shows that monetary policy is a bit 80% credible. That is, only about 20% inflation expectation is gotten in response to a 1 percentage change in inflation. A robustness check reveals that the degree of inflation expectations only increased by 1% to 21%. Examining the effect of anchoring of inflation rate and exchange rate on the real GDP and the sub-sectors’ outputs, an error correction model was employed. The choice of ECM approach was because all of the variables were stationary at first difference; most of the models were cointegrated and monetary policy is more effective in the short-run. The result on the aggregate output model is presented in Table 5. The model has a strong goodness of fit and there is no suggestion of serial correlation as the Durbin Watson is within a reasonable bound. The coefficient of the error correction is rightly signed and it shows about 37.4% of the deviation from the long-run or equilibrium is corrected in the short-run.
Table 5. ECM Estimated Result on Output Equation
Dependent Variable: ∆GDP
Variable |
Coefficient |
Std. Error |
t-Statistic |
Prob. |
|
|
|
|
|
|
|
|
|
|
∆ (EXCH * IA)t |
-0.003735** |
0.001149 |
-3.251028 |
0.0044 |
∆ (EXCH*IA)t-1 |
-0.002895 |
0.001720 |
-1.683639 |
0.1095 |
∆LOG(GOVCAP)t |
-0.011575 |
0.012857 |
-0.900268 |
0.3799 |
∆LOG(GOVCAPt-1 |
0.030345 |
0.016484 |
1.840840 |
0.0822 |
∆LOG(GOVCAPt-2 |
-0.040347* |
0.017001 |
-2.373186 |
0.0290 |
∆LOG(INV)t |
0.150553** |
0.029478 |
5.107323 |
0.0001 |
∆LOG(INVt-1 |
-0.033762 |
0.028173 |
-1.198365 |
0.2463 |
∆LOG(INVt-2 |
0.085798** |
0.023635 |
3.630111 |
0.0019 |
∆LOG(CPS)t |
0.001587 |
0.028309 |
0.056068 |
0.9559 |
∆LOG(CPSt-1 |
-0.057941 |
0.030553 |
-1.896418 |
0.0741 |
ECMt-1 |
-0.373734* |
0.114378 |
-3.267541 |
0.0043 |
R-squared |
0.999333 |
|
|
|
Adjusted R-squared |
0.998778 |
|
|
|
Prob(F-statistic) |
0.000000 |
|
|
|
Durbin-Watson stat |
2.155550 |
|
|
|
* Significant at 5%
** Significant at 1%
Source: Authors’ Computation
The result in Table 5 further reveals a statistically significant indirect effect of exchange rate interaction with inflation anchoring on economic growth. This is in a way in line with the work of Audu & Amaegberi (2013) and Falana (2019) but contrary to the studies by Marshall (1923), Hicks (1937), Lerner (1944), Flemming (1962), Mundell (1963) and Speller (2006) on the nexus between exchange rate and growth. The implication of this result is that inflation anchoring in the midst of a depreciating exchange rate has contractionary effect on economic growth in Nigeria. However, government capital expenditure in the past two years, domestic investment in current year and in two previous years have expansionary effect on economic growth in Nigeria.
Table 6. ECM Estimated Result on the Five Sector Output Equations
Variable |
Agric |
Industry |
Building |
Trade |
Services |
|||||
Coefficient |
Prob |
Coefficient |
Prob |
Coeffici. |
Prob |
Coeffici. |
Prob |
Coefficient |
Prob |
|
|
|
|
|
|
|
|
|
|
|
|
∆ (EXCH *IA) |
0.002201 |
0.4897 |
-0.007340** |
0.0060 |
-0.000354 |
0.8666 |
- |
- |
-0.002906 |
0.0872 |
∆ (EXCH *IA)t-1 |
-0.017418** |
0.0008 |
-0.006796* |
0.0553 |
- |
- |
-0.00065* |
0.0212 |
-0.001577 |
0.5746 |
∆ (PLRt |
0.003869 |
0.1447 |
0.003770 |
0.1096 |
- |
- |
- |
- |
0.002159 |
0.1967 |
∆ (PLRt-1 |
- |
- |
-0.003268 |
0.2623 |
- |
- |
- |
- |
0.00370* |
0.0400 |
∆LOG(CPSt |
0.093708* |
0.0410 |
-0.010397 |
0.8575 |
0.07865* |
0.0414 |
0.1240** |
0.0034 |
0.06260* |
0.0010 |
∆LOG(CPSt-1 |
- |
- |
-0.129535* |
0.0326 |
- |
- |
- |
- |
- |
- |
∆GOVCAP) |
-0.041282 |
0.2288 |
-0.008647 |
0.5888 |
-0.011939 |
0.6905 |
-0.014581 |
0.6374 |
- |
- |
∆logGOVCAPt-1 |
0.113141** |
0.0069 |
- |
- |
- |
- |
- |
- |
- |
- |
∆LOG(INV)t |
- |
- |
- |
- |
0.135556 |
0.0572 |
0.024712 |
0.7241 |
0.040162 |
0.3397 |
∆LOG(INDt-1 |
- |
- |
0.169422 |
0.2010 |
- |
- |
- |
- |
- |
- |
∆LOG(BUILD)t-1 |
- |
- |
- |
- |
0.244338 |
0.1147 |
- |
- |
- |
- |
∆LOG(TRADEt-1 |
- |
- |
- |
- |
- |
- |
0.407** |
0.0191 |
- |
- |
ECMt-1 |
-0.568023** |
0.0001 |
-0.905824** |
0.0000 |
-0.3843** |
0.0008 |
-0.24064* |
0.0401 |
-0.2944** |
0.0114 |
R-squared |
0.994644 |
|
0.984963 |
|
0.991620 |
|
0.996343 |
|
0.998921 |
|
Adj. R-squared |
0.992716 |
|
0.976761 |
|
0.989825 |
|
0.995027 |
|
0.998274 |
|
Prob(F-statistic) |
0.000000 |
|
0.000000 |
|
0.000000 |
|
0.000000 |
|
0.000000 |
|
D-Watson stat |
2.306083 |
|
1.860733 |
|
2.053602 |
|
2.131892 |
|
2.296231 |
|
* Indicates significant at 5%
** Indicates significant at 1%
Source: Authors’ Computation
The results on the sub-sectors are presented in Table 6. The models have strong goodness of fit and their Durbin Watson statistics are within reasonable bounds. The coefficient of the ECM of the agricultural sector’s model is correctly signed. Approximately, 56.8 percent deviation from the long-run was corrected in the short-run. There is a statistically significant contractionary impact of exchange rate interaction with inflation anchoring on previous year agricultural output growth. This is somehow consistent with the work of Yaqub (2010) and Falana (2019). The result also shows that first lagged government capital expenditure and credit to private sector have a positive and significant impact on agricultural output which conform to a priori expectations. The ECM of the industrial sector model is rightly signed. Almost 90.58 percent deviation from equilibrium was rectified in the short- run. There is an indirect significant impact of exchange rate interaction with inflation anchoring on industrial output growth. Like agriculture, the result is somehow consistent with the findings of Yaqub (2010) and Falana (2019). The building and construction model also have a correctly signed ECM. Approximately, 38.43 percent deviations from long-run equilibrium was rectified in the short-run. There is an insignificant indirect impact of exchange rate interaction with inflation anchoring on building and construction output growth. Besides, the trade and service models have correctly signed ECMs such that about 24.1 percent and 29.4 percent deviations, respectively, from long-run were rectified in the short-run. There is a statistically significant indirect impact of exchange rate interaction with inflation anchoring on trade output growth which someway supports the finding of Falana (2019). However, the impact of exchange rate interaction with inflation anchoring on service output growth is statistically insignificant.
Table 7. Spline Regression Estimated Result
Dependent Variable: GDP
Variable |
@6% threshold |
@7% threshold |
@8% threshold |
@9% threshold |
@10% threshold |
|||||
Coefficient |
Prob |
Coefficient |
Prob |
Coeffici. |
Prob |
Coeffici. |
Prob |
Coefficient |
Prob |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
C |
23.76190 |
0.0000 |
24.02060 |
0.0000 |
23.81762 |
0.0000 |
23.86924 |
0.0000 |
23.89513 |
0.0000 |
IA*EXCH |
0.001131 |
0.5561 |
0.001391 |
0.4657 |
0.001095 |
0.5682 |
0.001233 |
0.5271 |
0.001295 |
0.5074 |
INF |
-0.000582 |
0.4700 |
-0.001069 |
0.1913 |
-0.000927 |
0.3069 |
-0.000506 |
0.5716 |
-0.000452 |
0.6097 |
(INF-K) |
0.019665 |
0.6644 |
-0.041811 |
0.2638 |
-0.016377 |
0.6514 |
-0.012435 |
0.0005 |
-0.015734 |
0.0000 |
LOG(GOVCAP) |
-0.079258 |
0.0041 |
-0.078229 |
0.0037 |
-0.077594 |
0.0060 |
-0.082777 |
0.0029 |
-0.082471 |
0.0027 |
LOG(INV) |
0.102026 |
0.0019 |
0.095586 |
0.0029 |
0.101555 |
0.0019 |
0.099097 |
0.0024 |
0.098287 |
0.0026 |
LOG(CPS) |
0.233176 |
0.0000 |
0.229987 |
0.0000 |
0.230487 |
0.0000 |
0.235503 |
0.0000 |
0.234993 |
0.0000 |
R-squared |
0.987140 |
|
0.987594 |
|
0.987147 |
|
0.987131 |
|
0.987184 |
|
Adjs R-squared |
0.984568 |
|
0.985113 |
|
0.984577 |
|
0.984558 |
|
0.984621 |
|
Prob(F-statistic) |
0.000000 |
|
0.000000 |
|
0.000000 |
|
0.000000 |
|
0.000000 |
|
D-Watson stat |
1.772950 |
|
1.902140 |
|
1.837733 |
|
1.786127 |
|
1.762730 |
|
* Implies significant at 5%
** Implies significant at 1%
Source: Authors’ Computation
On the threshold result in Table 7, at an arbitrary threshold of 6%, the adjusted impact of inflation rate (INF-K) on output was positive but when it was increased to 7%, it became negative nonetheless statistically insignificant. The insignificance was also the case at 8% but at 9% threshold, the impact became negative and statistically significant even beyond the level. Therefore, the inflation threshold beyond which it becomes detrimental to the Nigerian economy is 9 percent. This is more realistic compared to 13 percent found in the study by Bawa & Abdullahi (2011) on the Nigerian economy.
5. Conclusion and Recommendations
In this paper, we have examined inflation anchoring, exchange rate and sectoral economic growth in Nigeria. Inflation anchoring with a rising exchange rate would have contractionary effects on economic growth, agricultural output, industrial output and trade output in Nigeria. Unfortunately, inflation anchoring with a depreciating exchange rate would not make significant effect on the building and construction output as well as service output in Nigeria. Therefore, this study has laid credence on the need to develop the agricultural and the industrial sectors of the Nigerian economy so that where inflation is anchored and exchange rate is at the same time devalued, they can promote the growth of outputs in the country. The agricultural and industrial sectors must continue to be promoted by the Nigerian government and its establishments including the Central Bank of Nigeria through provisions of low interest rate facilities. The loans must be disbursed in piecemeal to monitor performance and ensure effective utilization. Also, every state government must extensively delve into mechanized farming while emphasis should be on broadening of the agricultural value chain. As well-known from the economic literature, exchange rate devaluation would only be beneficial to the economy if the production base of the economy is strengthened. As enunciated above, there must be appropriate lending policies and government must rapidly rejuvenate the soft and hard infrastructures, especially institute efficient transport system, functional education and health system, clean water system and serious maintenance of law and order. Besides, the merchandize business which is flourishing in Nigeria must be developed further to promote economic growth in Nigeria but emphasis must be placed more on production than trading. The required development is in terms of trade which have to shift from foreign goods to locally made goods except for goods that Nigeria does not have comparative advantage. In addition, the monetary authority should unify its exchange rate system as the multiple exchange rate system has been compromised, hence not achieving its desired objectives.
The inflation threshold for Nigeria has been found to be 9 percent. Therefore, while it is exigent for Nigerian government to expand the economy, a hybrid inflation anchoring policy needs to be pursued. By implication, the monetary authority must not only target a single-digit inflation rate through a tight monetary policy, it must also extend credit to needed sectors of the economy that would promote economy growth and create jobs. This is not a contradictory approach as could be implied by conservative monetary policy but a way of simultaneously achieving low inflation and economic growth.
References
Audu, N. P. & Amaegberi, M. (2013). Exchange rate fluctuation and inflation targeting in an open economy: Econometric approach. European Journal of Accounting, Auditing and Finance Research, Vol. 1, no. 3, pp. 24-42.
Ball, L. & Mazumder, S. (2011). Inflation dynamics and the great recession. Brookings Papers on Economic Activity, pp. 337-406.
Barro, R. J. (1995). Inflation and economic growth. National Bureau of Economic Research (NBER), Working Paper no. 5326.
Bawa, S. & Abdullahi, I. S. (2011). Threshold effect of inflation on economic growth in Nigeria. CBN Journal of Applied Statistics, Vol. 3, no. 1, pp. 43-63.
Cecchetti, Ss. & Hakkio, C. (2009). Inflation targeting and private sector forecasts. National Bureau of Economic Research Working Paper no w15424.
Choi, S.; Fluceri, D. & Loungani, P. (2018). Inflation anchoring and growth: Evidence from sectoral data. IMF Working Paper WP/18/36.
Cruijsen, C. & Demertzis, M. (2011). How anchored are inflation expectations in EMU Countries? Economic Modelling, Vol. 1-2, pp. 281-298.
Demertzis, M. & Viegi, N. (2009). Inflation targeting: A framework for communication. The B. E. Journal of Macroeconomics, Vol. 9, no. 1, pp. 1-30.
De Pooter, M.; Robitaille, P.; Walker, I. & Zdinak, M. (2014). Are inflation expectations well anchored in Brazil, Chile and Mexico? International Journal of Central Banking, Vol. 10, no. 2, pp. 337-400.
Dickey, D. A. & Fuller, W. A. (1981). Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica, Vol. 49, pp. 1057-1072.
Dornbush, R., Fischer, S. & Kearney, C. (1996). Macroeconomics. Sydney, The Mc-Graw Hill Companies, Inc.
Engle, R. F. & Granger, C. W. J. (1987). Co-integration and error Correction: Representation, estimation, and testing. Econometrica, Vol. 55, pp. 251-276.
Falana, O. E. (2019). Exchange rate fluctuation and real sector performance in Nigeria: A disaggregated analysis. European Scientific Journal, Vol. 15, no. 25, pp. 358-386.
Fischer, S. (1993). The role of macroeconomic factors in growth. National Bureau of Economic Research (NBER), Working Paper no. 4565.
Flemming, M. (1962). Domestic financial policies under fixed and under floating exchange rates. International Monetary Fund Staff Papers, Vol. 9, pp. 369-379.
Gondo, R. & Yetman, J. (2018). Anchoring of inflation expectations in Latin America, Serie de Documentos de Trabajo. Working Paper Series.
Hattori, M. & Yetman, J. (2017). The evolution of inflation expectations in Japan. Journal of the Japanese and International Economies, Vol. 46, pp. 53-68.
Hansen, A. H. (1953). A Guide to Keynes. New York, Mc-Graw Hill.
Hicks, J. R. (1937). Mr. Keynes and the ‘Classics’: A suggested interpretation. Econometrics, Vol. 5, pp. 147-159.
Johansen, S. (1991). Estimation and hypothesis testing of cointegration vectors in Gaussian vector vector autoregressive models. Econometrica, Vol. 59, pp. 1551-1580.
Khan, M. S. & Senhadji, M. A. (2001). Threshold effects in the relationship between inflation and growth. IMF Staff Papers, Vol. 48, no. 1.
Kozicki, S. & Tinsley, P. A. (2012). Effective use of survey information in estimating the evolution of expected inflation. Journal of Money, Credit and Banking, Vol. 44, no. 1, pp. 145-169.
Lerner, A. (1944). The Economics of Control. New York: Macmillan.
Maddala, G. & Kim, I. (1998). Unit Roots, Cointegration and Structural Change. Cambridge, Cambridge University Press.
Marshall, A. (1923). Money, Credit and Commerce. London, Macmillan and Co.
Mehrotra, A. & Yetman, J. (2017). Decaying expectations: What inflation forecasts tell us about the anchoring of inflation expectations. International Journal of Central Banking, forthcoming.
Mordi, C. N.; Adebiyi, M.A.; Adenuga, A. O.; Adebayo, O. M.; Abeng, M. O.; Akpan, I. N. & Evbuomwan, O. O. (2013). Modelling the Real Sector of the Nigerian economy. CBN, Research Department, pp. 1-59.
Mundell, R. (1963). Inflation and interest rate. Journal of Political Economy, Vol. 71, no. 3, pp. 280-283.
Muth, J. F. (1961). Rational expectations and the theory of price movements. Econometrica, Vol. 29, no. 3, pp. 315-335.
Ncube, M. & Ndou, E. (2011). Inflation targeting, exchange rate shocks and output: Evidence from South Africa. African Development Bank Group Working Paper No. 134.
Obstfeld, M. & Rogoff, K. (1995). Exchange rate dynamics redux. Journal of Political Economy, Vol. 103, pp. 624-660.
Solow, R. (1956). A contribution to the theory of economic growth. Quarterly Journal of Economics, Vol. 70, pp. 65-94.
Speller, W. R. (2006). Real Exchange Rate in the Industrialized Commodity Currency Economies: An Error-Correction Framework. Unpublished Master’s Thesis, Stockholm School of Economics, Sweden.
Stockman, A. C. (1981). Anticipated inflation and the capital stock in a cash-in-advance economy. Journal of Monetary Economics, Vol. 8, pp. 387-393.
Strohsal, T. & Winkelmann, L. (2015). Assessing the anchoring of inflation expectations. Journal of International Money and Finance, Vol. 50(C), pp. 33-48.
Swan, T. W. (1956). Economic growth and capital accumulation. Economic Record, Vol. 32, no. 334-361.
Tobin, J. (1965). Money and economics. Econometrica, Vol. 33, no. 4, pp. 671-684.
Yaqub, J. O. (2010). Exchange rate changes and output performance in Nigeria: A sectoral analysis. Pakistan Journal of Social Sciences, Vol. 7, no. 5, pp. 380-387.
1 Senior Lecturer, Department of Economics, University of Lagos, Nigeria, Address: University of Lagos, University Road Lagos Mainland Akoka, Yaba, Lagos, Nigeria, Corresponding author: okdam76@yahoo.com.
2 Erstwhile Postgraduate Student, Department of Economics, Faculty of Social Sciences, University of Lagos, Nigeria, Address: University of Lagos, University Road Lagos Mainland Akoka, Yaba, Lagos, Nigeria, E-mail: augustus4success@yahoo.com.
AUDOE, Vol. 17, No. 1/2021, pp. 192-208