Abstract
Nigerian
agricultural economy was characterized by low capacity utilization of resources
that was further worsened by declined allocation functions and poor performance
of Major agricultural facilities. Monetary authorities were unable to
stimulate, direct and stabilize agricultural sector performance. Given the
prolonged food crisis in Nigeria, it was not known whether the observed
performance of the agricultural sector should be attributed to either
inefficient monetary policy or technical inefficiency. Secondary data collected
were reviewed, lagged and analyzed. The hypothesis was tested using
F-statistics and T-test.
The outcome of the test of hypothesis showed that the
observed F-value (54.47955*) was statistically significant at all distributions
(F* > F0.01), hence the conclusion that monetary policy did influence
farm outputs and the overall agricultural sector performance. Monetary
aggregates shifted resources out of agricultural sector and the results showed
that farm output and income velocity moved in opposite directions, hence
improvements in farm outputs and income velocity were not able to defray
inflationary pressures. This provided a strong reason why banks were reluctant
to lend to farmers. Money broadly defined (M2) shifted investment
portfolios out of agriculture and the impact was statistically significant at
1% level.The incidence of persistent low cash reserve requirement increased
bank lending rates volatility.
Keywords:Monetary
aggregates, economic indicators, farm outputs, money narrowly and broadly
defined, trade and non-trade sectors.
Introduction
Nigerian economy was characterized by
poor performance of the agricultural sector and government commitment in terms
of budgetary expenditure on agriculture per fiscal year over the studied
period, was insignificant when compared to aggregate government expenditure.
The economy also experienced ineffective distribution functions, declining
agricultural capacity utilization and poor performance of major agricultural
facilities. Nigerian economy grossly displayed a weak agricultural base and low
quality physical and social services (Garba,2000Ebo, 1995; Evbuomwan, 1996 2000
and CBN, 2000c).
Economic policy is a course of action
intended to achieve or mitigate socioeconomic problems (McConnel and Brue,
1999). Resources flow and allocation to the agricultural sector were highly
influenced by monetary policies. Monetary policy actions in Nigeria varied
considerably between 1970 and 2010. Monetary Authorities in Nigeria were unable
to stimulate, direct and stabilize agricultural sector performance both at
short and long-run (Ajayi and Ojo, 1981; Aham, 1992 and Afolabi, 1998 ).
Sustainable consolidation in agriculture was difficult to achieve in Nigeria
because public policy options failed to address fundamental problems of the
rural farmers (Mark, 1999; CBN, 2000a). Effort made to pursue economic policy
measures that can capture and improve the performance of the agricultural
sector failed, generating profound policy implications (Oran et al, 1979;
UNFPA, 1986; IRRI, 1988 and Ahan, 1992). Hence, basic agricultural policy
problems were in existence over the period and the policy regimes were such
that showed persistent long-term profile of policy interruption,
discontinuation and volatility making agricultural sector prone to both
empirical investigation and distributive judgment (Drabenstott and Tim, 1996;
Colander, 1998).
The key policy questions therefore were
centered on whether farm outputs were in harmony with monetary policy intents
and purpose? The structure and conduct of the policy actions measured in terms
of emphasis, policy mix and commitment determined the efficiency of the policy
instruments. Government had the aptitude and latitude to manipulate selected
monetary aggregates in pursuance of her policy objectives and often without
knowing in advance the precise impact on the desired targets. Given the
prolonged Nigeria’s food crisis and the insignificant agricultural share of the
nation’s GDP and exports, it was not known whether the observed performance
indices of the agricultural sector should be attributed to inefficient and
misconceived monetary policies and or inefficient agricultural resources
allocation and utilization. Hence, this work was intended to capture the
response of farm outputs to monetary policies in Nigeria and to recommend
appropriate policy options that can reposition agricultural sector gainfully in
the scheme of both domestic and international commodity markets. Therefore, it
was possible to hypothize that monetary aggregates and economic indicators did
not influence farm outputs and the overall performance of the agricultural
sector in Nigeria. To enhance agricultural sector performance, monetary
policies that discouraged under employment and or low capacity utilization of
agricultural resources were necessary (Lipton, 1977; Gardner, 1981; Dennis and
Alfred 1997; Mark, 1999 and McCallum, 2000).
The results provided an in-depth policy
options for modeling agriculture into future policy process in Nigeria
.The policy options offered were such
that have the potency to drive agricultural productivity forward and turn
Nigerian agricultural economy into market-oriented. (Garba,2000; Evbuomwan,
1996 2000 ; CBN, 2000c and Tweeten, 1980, 1983).
Methodology
It is imperative to investigate and
capture the farm outputs response to monetary policy mix but given the research
period of four decades (1970-2010) the study involved the collection and
analysis of large volume of data of diverse nature and sources. These informed
the purposive choice of crop sub-sector for
the study and as the index of analysis (CBN, 2000b ; McClave and
Sincich,2000).
Food production took considerable time
each year and farmers’ output adjustment with respect to monetary aggregates
and economic indicators adjustments were not instantaneous. This implied that
farm outputs supply adjustments were perceptible in the commodity markets only
after a period not less than one year (Henderson and Qundt, 1980). Therefore,
the quantity of farm output demanded in a particular year was a function of the
price in the said year but the quantity supplied in the same said year was a
function of the price in the previous year. Therefore, the lagged values of the
exogenous variables that took into account, the length of time in the overall
adjustment process of economic aggregates were used for regression analysis
(Christopher, 1992). The use of the lagged values of the variables was the most
efficient method to render economic behaviours dynamic and as partial
adjustment mechanism, it reflected the actual change in farm outputs in the
period (t). Therefore the response of farm outputs to lagged monetary
aggregates was implicitly expressed as:
Yht
= F (Xt-1, Xt-2, ……….., Xt-12) + U …… Eq(1)
Where
Yht was farm output contributions to GDP measured in tonnes in the
period (t) and (F) was the functional relationship and (U) was the error term.
Then, Xt-1, Xt-2, ……….., Xt-12 were money
narrowly defined (M1), money broadly defined (M2),
broader monetary aggregate (M3), cash reserve requirement, Discount
rates, exchange rates, banks’ CBN balances, lending interest rates, banks subscribed
government bonds, inflation rates, agricultural credit supplied and banks
liquidity ratio respectively. The data was collected from the Central Bank of
Nigeria and Federal Bureau of Statistics (CBN, 2010; FBOS, 2010).
Equation
(1) was explicitly expressed as:
a.
Linear
function
Yht=
a + b1 X t-1 + ………………… + b12 Xt-12
+ U …. Eq(2)
b. Semi-log
function
Yht=
a + b1 logX t-1 + …………… + b12 log Xt-12
+ U …. Eq(3)
c. Double-log
function
logYht= a + b1 logX
t-1 + …………… + b12 log Xt-12 + U …. Eq(4)
Where
the a priori expectations were b1< 0, b2<0, b3>0,
b4<0, b5<0, b6<0, b7>0,
b8<0, b9<0, b10<0, b11>0
and b12>0 and (a) was the intercept while the b5 were
the parameter estimates of the independent variables (John and David, 1986).The
hypothesis was tested using F-statistics and T-test (Christopher,1992 ; Murray
and Larry, 1999 ;Kerry, 2000; Rangaswamy, 2007 and Upender, 2008)
Results and
Discussion
The new innovations in global monetary
economics steered up various economic reforms and adjustments in Nigeria.
Monetary Authorities in Nigeria therefore, influenced the economy, using
monetary policy instruments to stimulate, stabilize and direct resources
endowment flow and growth. Major policy targets that were captured included
production, consumption, income distribution and trade among others. In
Nigeria, various economic sectors and units were differently captured by
government policy actions and in most cases, the outcome were not in harmony
with government policy intent and directions. Hence, monetary policies were
strongly believed to have shifted resource out of the agricultural sector in
Nigeria. Farm outputs appeared highly volatile, prompting such key policy
questions as whether agriculture did receive adequate prudent policy attention.
Hence this research effort and the results that follows, as shown in Table, 1.
Table, 1: Goodness of
fit Statistics of Farm output Response to Monetary Aggregates and Economic
Indicators
Parameters lin(Eq.2) lin-log(Eq.3) log-log(Eq.4)R 0.98651 0.99088 0.98540
R2 0.97320 0.98184 0.97102
(97%) (98%) (97%)
R-2 0.95534 0.96974 0.95170
(96%) (97%) (95%)
SEE 6931.09 5705.20 0.13098
(2.7%) (2%) (3%)
DW 1.62311 2.20964 2.32094
Observed F-value 54.47955* 81.12112* 50.25456*
Sig. F-value 0.0000 0.0000 0.0000
Regression DF 12 12 12
Residual DF 18 18 18
Tabular
F-value(1%) 3.37 3.37 3.37
Source: Estimates
from multiple regression analysis of field data, 2012.
Notes: *
= significant at 1% level of confidence.
In Table 1, the goodness of fit
statistics of output response to monetary aggregates and economic indicators
was presented in lin (Eq.2), lin-log (Eq.3) and log-log (Eq.4). In lin-log
(Eq.3) the explanatory power (98%) was ample and standard error estimates (2%)
inconsiderable relative to the parameters of lin(Eq.2) and log-log (Eq.4). The
size of the adjusted R-Square (97%) obtained for lin-log (Eq.3) implied that
the explanatory power (R2) of 98% was not magnified. However, in
conducting the test of the null hypothesis and subsequent discussions of the
results, lin (Eq.2) was chosen based on its consistency with plausible
econometric attributes and it gave in addition, a good fit. The choice of lin
(Eq.2) for the test of hypothesis was further prompted by its greater number of
variables that were statistically significant and consistent with a priori
expectations. Its explanatory power (97%) was also considered profuse and the
standard error estimate (2.7%) was infinitesimal. The adjusted R-Square of about
96% confirmed that the R-Square was not exaggerated (Christopher, 1992;
Upender, 2008; Kerry, 2000; Murry and Larry, 1999 ;Rangaswamy,2007).
The general direction of change in the
regression analysis of lin (Eq.2) was an increase since the intercept was positive.
The goodness of fit statistics in Table 1, showed that farm outputs response to
the combined impact of monetary aggregates and economic indicators exhibited a
linear supply function. The multiple correlation coefficient (0.98651) and the
R-Square (0.97320) were statistical estimates subject to error and were
therefore tested for reliability and degree of confidence (Table, 1). The
outcome of the test showed that the observed F-value (54.47955*) was
statistically significant at 1% level of confidence, contrary to the
theoretical assumptions. Specifically, the F-value was significant at 0.0000,
which implied that it was statistically significant at all distributions. Hence
the rejection of the null hypothesis that monetary aggregates and economic indicators
did not influence farm outputs and the overall performance of the agricultural
sector in Nigeria.
The
alternative hypothesis was therefore accepted with the understanding that there
was only one chance per hundred that the null hypothesis would be accepted when
it should be rejected. Thus there was 99% confidence that rejecting the null
hypothesis was a right decision and that the acceptance of the alternative
hypothesis was right.
The
Durbin-Watson d-statistics test and multicollinearity diagnostics test
conducted did not show any strong evidence of autocorrelation and
multicollinearity respectively. Equation (2) showed the parameters and standard
error estimates of variables specified to test farm outputs response to
monetary aggregates and economic
indicators as follows:
Yht = 419 – 0.0001Xt-1
– 0.081Xt-2 + 0.082Xt-3*** + 22X+t-4
– 244Xt-5 – 261Xt-6 – 1.12Xt-7 – 279Xt-8
-
(963)
(0.162) (0.15)
(0.048) (252) (179)
(620) (0.674) (123)
0.714Xt-9*** + 81Xt-10 +
4.47Xt-11* + 227Xt-12***
(0.37)
(112) (1.47) (127) …… Eq. (2)
R2
= 0.97320 (97%), R-2 = 0.95534, SEE = 6931.09 (2.7%), DW = 1.62311,
F-value = 54.47955*,
*
= Significant at 1%, ** = Significant at 5%, *** = Significant at 10% level of
Confidence.
Where
b1<0, b2<0, b3>0, b4<0,
b5<0, b6<0, b7>0, b8<0,
b9<0 b10<0, b11>0,and b12>0.
The
intercept of lin (Eq.2) was positive but not statistically significant. This
implied that the direction of change in the regression was an increase and that
the average value of farm output contributions to GDP when the explanatory
variables were set at zero was 419 tonnes of farm outputs.
The
negative correlation (-0.001) of money narrowly defined (Xt-1)
conformed to the theoretical expectations. This implied that farm output and
money narrowly defined moved in opposite directions. The impact of money
narrowly defined on farm outputs contributions to GDP was not statistically
significant but farm output performance was more consistent compared to the
supply of money narrowly defined (M1). The money narrowly defined (M1)
was inelastic with respect to changes in farm output. The negative correlation
of farm output response to changes in money narrowly defined (M1)
implied that income velocity and farm output did not increase or decrease
simultaneously in the same direction. Thus improvements in income velocity and
farm output were infinitesimal to defray inflationary pressures and this
provided a strong evidence why commercial banks were reluctant to lend to
farmers as farmers were discouraged from borrowing from banks. This result
therefore confirmed the general low investments in agricultural sector across
the research period.
The M2, money broadly defined
(Xt-2) and M3 - broader monetary aggregates (Xt-3)
were expectedly in conformity with a priori expectations. The M2 was
statistically insignificant while M3 was statistically significant
at (F*>F0.10). Money broadly defined changed simultaneously in
opposite direction with respect to farm output while broader monetary
aggregates changed in the same direction. The negative correlation coefficient
of money broadly defined (M2) was not statistically significant but
it shifted investment portfolios from the high risky, less liquid and less
profitable agricultural sector to other lucrative sectors. Banks invested
immensely on government bonds and Treasury bills that were more profitable
compared to investments in agriculture.
Farm
output was more consistent with respect to money supply (M2 and M3)
but money supply elasticity of farm output was inelastic. Broader monetary
aggregate (M3) gave the overall liquidity of the economy and its
impact was statistically significant at (F*>F0.10) level.
The cash reserve requirement ratio (Xt-4)
was positively correlated with farm output contrary to a priori expectations
but its impact was statistically insignificant. Cash reserve requirement ratio
elasticity of farm output was inelastic but farm output was more consistent
compared to cash reserve requirement. Cash reserve requirement appeared to have
stimulated bank lending to agriculture but it complicated monetary policy
operations by removing distortion tax on depository banks as low cash reserved
requirement ratio prevailed and consequently increased bank lending rates
volatility. The cash reserve requirement ratio differed behaviorally with the
CBN bank balances (Xt-7) which unexpectedly was negatively
correlated with farm output contrary to theoretical expectations. Its impact
was statistically insignificant but it created two policy problems. Though farm
output was consistent with respect to CBN bank balances and the bank balances
elasticity of farm output was inelastic. Bank balances arising from
recapitalization created long term adverse effect following its spurious
relationship with other monetary aggregates. Accordingly, this work agreed with
Sellon and Weiner (1997) that bank recapitalization arise from payment needs
and timing rather than a mandated linkage to deposit liabilities. This
consequently, re-engineered the payment system structure as an important factor
in monetary policy formulation and implementation. Thus changes in the payment
system affected the demand for settlement balances and complicated monetary
policy actions. Traditionally, CBN could have lowered the cash reserved
requirement ratio to avoid limiting the volume of investment portfolio to less
liquid agricultural sector and thus expand aggregate money supply.
Contrary
to theoretical expectations, discount rate (Xt-5) was positively
correlated to farm outputs. This appeared to suggest that CBN encouraged and
eased money and agricultural credit supply to enhance farm outputs. Discount
rate elasticity of farm output was inelastic and it had greater variability
compared to farm outputs. Changes in discount rates were not supported by
appropriate changes in monetary policy actions; hence the changes did not make
impacts that were statistically significant. In conformity with a priori
expectations, the exchange rate (Xt-6) regimes were negatively
correlated to farm outputs. The exchange rates impacts were statistically
insignificant and volatile. The response of farm output to exchange rates was
inelastic but more consistent. The magnitude of exchange rate negative
correlations adversely affected agricultural production and trade volumes.
Nigerian economy experienced positive excess demand for foreign exchanged
arising from overvalued domestic currency, hence trade in goods and non-factor
services were negative. Consequently, this constrained Nigeria to over draw
from her foreign reserves and then turned to borrow from world capital markets
to maintain balance of trade. The result was low opportunity cost of resources
employed in Nigeria’s rural economy and preference of Nigerians to place more
savings in foreign assets. The situation would have been worse, if the impact
of the exchange rate was statistically significant.
Expectedly
and in conformity with a priori expectations, the bank lending rate (Xt-8)
had negative correlation with farm outputs. That is, when lending rates
increased, farm outputs decreased and vice versa. The impact of lending rate on
farm output was not statistically significant and farm output had greater
variability compared to lending rates. Banks lending rate
elasticity of farm output was less than unity. Banks subscribed government
bonds (Xt-9) had negative correlation with farm output and was thus,
in harmony with theoretical expectations. The impact of bank subscribed
government bonds on farm output was significant at 10% but it had greater
variability compared to farm outputs. Government bonds elasticity of farm
output was inelasticity. However, its negative correlation implied that banks
shifted their portfolio investments from agricultural financing to highly
liquid government securities and bonds. Contrary to theoretical expectations,
inflation rate (Xt-10) was positively correlated with farm outputs,though
its impact was not statistically significant. Farm output was however more
consistent with respect to inflation rate. Inflation rate elasticity with
respect to farm output was less than unity.
Agricultural
credit supply (Xt-11) had positive correlation with farm output and
this was in conformity with a priori expectations. The impact of agricultural
credit supply on farm outputs was statistically significant at 1% level but
farm output was more consistent with respect to agricultural credit supplied.
Agricultural credit supply elasticity of farm output was about unity. That is,
banks expanded agricultural credit supply proportionally by the same percentage
change in farm outputs. Banks liquidly ratio (Xt-12) had positive
correlation with farm output in conformity with theoretical expectations. Its
impact on farm output was statistically significant at (F*<F0.10)
level but banks liquidity elasticity of farm output was less than unity.
Conclusion
Significant share of the adjustments
burden created by volatile monetary aggregates and economic indicators were
borne by the agricultural sector. Given the prevailed exchange rates regime,
changes in monetary policies had significant policy implications that were
reflected in changes in real exchange rates and not in real interest rates.
This policy implication was a burden of adjustment passed to the trade sectors
which included export sectors and those sectors such as agriculture that
competed with imports (CBN 2000a, Barkema and Doye 1985, Falk et al, 1986).
Nigeria’s agricultural sector was a
trade sector, given the nation’s food imports and exports of agricultural
commodities. In this context, the results showed that adjustments in monetary
policies shifted to non-trade activities. Therefore, the farmers’ supply
function in Nigeria showed precisely the response of farm outputs to monetary
aggregates volatility and how farmers adjusted their farm outputs to the
prevailing food prices (Ojwang, 1996).Therefore, Nigerian agricultural sector
should be restructured to oligopolistic economy and government distribution
function policy overhauled to ensure distributional equity.
References
Afolabi,
L. (1998) Monetary Economics, Lagos, Perry Barr Ltd. P13-15.
Aham,
A. (1992) The Nigerian Economy Under Structural Adjustment Programme, P5-14
Nigerian Economy: Issues and
Trends, Enugu, Acafor Publishers.
Ajaji,
S.I. and Ojo, O. (1981) Intermediate Macroeconomics. Ife University of Ife
Press. P169-170.
Barkema,
A.D. and Doye, D.G. (1985) Farm Survival Under Financial Stress, IOWA, IOWA
State University. p6-16.
Central
Bank of Nigeria (2000a) Annual Report and Statement of Accounts for 1999,
Abuja, CBN. P1-158.
Central
Bank of Nigeria (2000b) Statistical Bulletin, 11(2): 1-150.
Central
Bank of Nigeria (2000c) The Changing Structure of the Nigeria Economy and
Implications for Development, Abuja, CBN, P37-53.
Central
Bank of Nigeria (2010) Statistical Bulletin, 13(4): 1-172
Christopher,
D. (1992) Introduction to Econometrics, New York, Oxford University Press. P143-166.
Coleander,
D.C. (1998)Macroeconomics, 3rd edition, New York, Irwin McGraw-Hill,
P334-357.
Dennis,
R.A and Alfred, J.F. (1997) International Economics New York, Irwin
McGraw-Hill, p203-352.
Drabenstott,
M. and Tim, R.S. (1996) The Changing Economy of the Rural Heartland, P1-11
Economic Economic
forces shaping the rural heartland,
Kansas city, federal reserve bank of Kansas city, USA.
Ebo,
E.C. (1995) Sustainable Development: the Theory and Implications for Rural
Nigeria, P3-12,Rurall development in Nigeria: concepts, process and
prospects, Enugu, Auto-century publishing company.
Evbuomivan,
G.O. (1996) An Empirical Analysis of the Prices of Nigeria’s Agricultural
Export Commodities. CBN Economic and Financial review. 34(1):501-512.
Falk,
B; Devadoss, S. and Meyer, W.H. (1986) Money, Inflation and Relative Prices:
Implication for USA Agriculture. Research working paper No 86WP. Centre for
Agricultural and Rural Development, IOWA State University.P1-29
Federal Bureau of Statistics (1999a) Annual
Abstracts of Statistics, Abuja Nigeria. P1-80.
Federal
Bureau of Statistics (2010) Annual Abstract of Statistics, Abuja Nigeria.
P1-63.
Garba,
P.K. (2000) An Analysis of the Implementation and Stability of Nigerian
Agricultural Policies , 1970-1993, Technical Report No 101, Nairobi Kenya,
African Economic Research Consortium, P34-43.
Gardner,
B. (1981) On the Power of Macroeconomic Linkage to Explain Events in USA
Agriculture. American Journal of Agricultural Economics, 63: 872-878.
Henderson,
J. M. and Qundt, R. E. (1980) Microeconomics Theory: A Mathematical Approach,
Singapore , McGraw-Hill Inc; P166-168.
International
Rice Research Institute (1988) Rice Facts, Philippine, IRRI. P1-5.
John,B.P
and David, A.I. (1986) Agricultural Finance: An introduction to micro and macro
concepts, Washington D.C. McGraw-Hill International, P373-526.
Kerry,
P. (2000) An Introduction to Applied Econometrics, New York, Palgrave
Publishers, P71-1183.
Lipton,
M. (1977) Why Poor People Stay Poor: Urban Bias in World Development, Cambridge
Mass, Harvard University Press. P25-80.
Mark,
D. (1999) Consolidation in USA Agriculture: The New Rural Landscape and Public
Policy, P63-71.
McCallum,
B.S.(2000) Theoretical Analysis Regarding a Zero Lower Bound on Nominal
Interest Rates, Journal of Money, Credit and Banking, 32(4):870-904.
McClave,
J.T. and Sincich, T. (2000) Statistcs, 8th edition, New Jersey,
Prentice Hall Inc; P514-517.
McConnel,
C.R. and Brue, S. I. (1999) Economics, New York , Irwin McGraw-Hill Publishers,
P857.
Murray,
S. and Larry, J.S. (1999) Statistics, New York , McGraw-Hill International.
P216-241.
Ojwang,
J.B. (1996) The Constitutional Basis for Environmental Management, P39-58. In
Land We Trust , Nairobi, Initiative Publishers.
Oram,
P; Juan, Z; George, A and Shyamal,R .(1979) Investment and Input Requirements
for Accelerating Food Production in Low-income Countries by 1990.IFPRI Research
Report, (10):1-169.
Rangaswamy,
R . (2007) Agricultural Statistics, 2nd edition ,New Delhi India,
New Age International (P) Ltd Publishers. P142-201.
Sellon,
G.H; Weiner, S.E.(1997) Monetary Policy Without Reserve Requirement : Case
Studies and Options for the USA. Federal
Reserve Bank of Kansas City, Economic Review, 82 (2):5-29.
Tweeten,
L.G. (1980) Macroeconomics in Crisis : Agriculture in an Underachieving
Economy. American Journal Agricultural Economics, 62: 5-29.
Tweeten,
L.G. (1983) Impact of Federal Fiscal-Monetary Policy on Farm Structures,
Southern Journal of Agricultural Economics, 15:61-68.
United
Nation Fund for Population Agency (1986) UNFPA Assistance and Population Data
1986 Activities, New York USA, UNFPA, P1-2.
Upender,
M. (2008) Applied Econometrics, 3rdedition , Pradesh India , Vrinda
Publications Ltd ,P1-295.