Question One
The purpose of this question is to apply the model proposed by Mundell, expand it and give
empirical analysis of the model. The model will be analyzed through an open economy, for the
country I choose South Africa and the analysis will involve a simple linear regression having one
dependent (GDP Growth Rate) and independent variable (Real Interest Rate) from the period
2011-2022. The period was considered because it had the available data to use in the analysis.
The country is in the Southern part of Africa and its among the giant economies of Africa. The
Mundell Fleming Model basically will be used to adds valuable depth to the Keynesian model of
macroeconomics by extending its analysis to include the foreign trade sector. Mundell Fleming
model assumes a world with 'perfect capital mobility' in an open economy whereby any small
differential in the domestic interest rate on an asset class will cause inflows/outflows of money
across the world in search of the highest returns, which would thereby equalize those returns
through the forces of supply and demand (www.dyingeconomy.com, 28 April 2022).
The following are the basic assumption of the model:
1) The domestic rate of interest (r) is equal to the world rate of interest (r*)
2) There is small open economy with perfect capital mobility.
3) It assumes fixed price level.
The main prediction from the Mundell-Fleming model is that the behaviour of an economy
depends crucially on the exchange rate system it adopts- whether it operates a floating exchange
rate system or a fixed exchange rate system.
The Open Economy IS Curve: In the Mundell-Fleming model, the market for goods and
services is expressed by the following equation:
Y = C(Y – T) + I(r*) + G + NX (e)
Where all the terms have their usual meanings. Here investment depends on the world rate of
interest r* since r = r* and NX depends on the exchange rate e which is the price of a foreign
currency in terms of domestic currency.
The Open Economy LM Curve: The equilibrium condition of the money market in the
Mundell-Fleming model is expressed as:
M = L(r*, Y) since r = r*.
Here the supply of money equals its demand and demand for money varies inversely with r* and
the positively with Y. In this model, M remains exogenously fixed by the central bank.
The Mundell Fleming Model with a Fixed Exchange Rate
As the graph illustrates, monetary policy under a fixed exchange rate regime is ineffective. The
reason for that is because any attempt by the government to boost the economy via a monetary
expansion will cause a fall in the domestic interest rate. With a lower interest rate, South African
financial assets offer lower returns than foreign financial assets, and this causes a capital outflow
from the domestic economy, which in turn creates a deficit on the balance of payments (BoP).
The central bank can try to plug the gap on the BoP by running down its reserves of foreign
currency, but with perfect capital mobility the outflow of foreign currency would be far larger
than the total amount of the central bank's reserves, and so any attempt at plugging the gap is
futile. The horizontal BP curve shows that only an interest rate of “R” gives stability on the BoP,
higher rates would give a surplus and a capital inflow, lower rates give a deficit and a capital
outflow (www.dyingeconomy.com, 28 April 2022).
Fiscal Policy
Whilst monetary policy is ineffective under fixed exchange rates, fiscal policy is very powerful.
If the government increases its spending in order to boost the consumption function, the result
can be illustrated as a rightward shift of the IS curve. Unlike monetary policy, the fiscal policy
expansion causes the interest rate to increase, which attracts an inflow of foreign money because
domestic capital assets now offer a higher yield. This causes a payments surplus on the BoP and
puts pressure on the domestic currency to revalue to a higher exchange rate. In order to maintain
the fixed exchange rate, the central bank is forced to expand the money supply in order to bring
the interest rate back down to its original level, so that domestic financial assets have the same
yield as before. This causes a rightward shift of the LM curve, and domestic output increases
from Y1-Y2 as illustrated. The increase in income/output here is in accordance with the full
effect of the Keynesian Multiplier.
The Model with a Floating Exchange Rate
In the Mundell Fleming model with a floating exchange rate, the central bank plays no role in
maintaining any particular value for the domestic currency, it simply allows supply and demand
to do its thing and settle at the equilibrium rate - the rate at which the BoP is in balance. Under
this system, when an expansionary monetary policy is enacted, the rightward shift of the LM
curve puts downward pressure on the interest rate just as before, but this time the central bank
does nothing when money starts to flow out of the economy in search of higher yields in foreign
markets. As a result, the exchange rate depreciates to a lower value, and this makes imports more
expensive and exports more competitive. Both of these effects improve the trade balance and
lead to an increase in consumption of domestic goods and services, which means that the IS
curve shifts to the right putting upward pressure on the interest rate.
At the point that the new IS curve intersects the new LM curve, the interest rate will return to its
original level stabilizing the inflow and outflow of capital such that the BoP is in equilibrium. At
the new equilibrium point, national income/output rises from Y1 to Y2, as shown in the diagram.
Monetary policy is therefore a very powerful tool in a floating exchange rate system with high
capital mobility which is precisely the opposite of what happens with fixed exchange rates. As a
quick qualifying point, you should note that at Y2, the domestic economy will maintain a lower
exchange rate, and therefore a higher level of exports and fewer imports. This will move the
current account of the BoP towards a surplus (or smaller deficit), but the capital account will
offset that with an equal move in the opposite direction - meaning reduced capital inflows into
the economy.
Fiscal Policy
Now consider a fiscal policy expansion. If the South African government increases its spending
in the economy to try and boost income/output, it will simply lead to an exchange rate
appreciation that offsets the policy effect. This adjustment process to the government spending
increase would first impact the interest rate, because the IS curve would shift to the right. The
higher interest rate would then cause foreign money to flow into the economy to purchase
domestic assets - thereby increasing domestic money demand to pay for those assets. The higher
demand for domestic currency would cause its exchange rate value to increase, and that would
cause a loss of competitiveness in trade. Imports would increase and exports would fall, causing
the IS curve to shift back to its original position. The only lasting effect of this fiscal expansion
is that the exchange rate is higher, and therefore the current account of the BoP will move
towards a deficit (or reduced surplus) whilst the capital account moves towards a surplus.
Income/output remains where it started (www.dyingeconomy.com, 28 April 2022).
Findings from the Analysis
Model Summaryb
Model
R
R Square
.285a
1
Adjusted R
Std. Error of
Square
the Estimate
.081
-.033
Durbin-Watson
2.73272
.461
a. Predictors: (Constant), Real Interest Rate
b. Dependent Variable: GDP Growth Rate
ANOVAa
Model
Sum of
df
Mean
Squares
Regression
1
F
Sig.
Square
5.294
1
5.294
Residual
59.742
8
7.468
Total
65.036
9
.424b
.709
a. Dependent Variable: GDP Growth Rate
b. Predictors: (Constant), Real Interest Rate
Coefficientsa
Model
Unstandardized Coefficients
Standardized
t
Sig.
Collinearity Statistics
Coefficients
B
(Constant)
Std. Error
-1.765
3.141
.662
.786
Beta
Tolerance
-.562
.590
.842
.424
VIF
1
Real Interest Rate
a. Dependent Variable: GDP Growth Rate
A simple linear regression model is stated below as follows
GDP = -1.765+ 0.662 Real Interest Rate1 + µ
Testing for heteroscedasticity
.285
1.000
1.000
Based on the scatter plot, the dots are far away from each other and that shows the presence of
heteroscedasticity in the variables and the histogram above does not clear show the pattern being
followed. Testing for multicollinearity will be based on the VIF value and if the VIF is greater
than 3 then there is presence of multicollinearity in the variables. The VIF value was less than 3,
therefore, it means that was no multicollinearity in the variables. Testing for autocorrelation will
be based on Durbin-Watson testing and if the value is less than 2 then it means that, there is
positive autocorrelation and if above 2, it means negative autocorrelation. The data above shows
that, there is a negative autocorrelation since the value is above 2.
The intercept coefficient value was -1.765 with the significance-value of 0.590. This shows that
the intercept was not significant and holding interest rate constant; GDP growth rate for South
Africa will reduce by 1.765. The interest rate coefficient value was 0.662 with the significancevalue of 0.424. This shows that interest rate not significant and with 1 percent increase in interest
rate, the GDP growth rate for South Africa will increase by 42.2 percent. Therefore, it can be
concluded that, high interest rates raises the cost of borrowing and thus reduce disposable
income, hence causes an appreciation in the exchange rate. The coefficient does not make any
economic meaning.
The significance of the regression model was not of good fit since the significant-F was greater
than the 0.05. The coefficient of determination was 0.081, meaning that, 8.1 percent of the data
was captured by the regression line and remaining percentage was not captured by the regression
line. Multiple R showed a positive linear relationship results since the value was 0.285 or 28.5%.
Question Two
Model Summaryb
Model
R
R Square
.956a
1
Adjusted R
Std. Error of the
Square
Estimate
.913
.870
.96853
a. Predictors: (Constant), Unemployment Rate, Real Interest Rate,
Inflation Rate
b. Dependent Variable: GDP Growth Rate
ANOVAa
Model
Sum of Squares
Regression
1
Residual
Total
df
Mean Square
59.407
3
19.802
5.628
6
.938
65.036
9
F
Sig.
21.110
.001b
a. Dependent Variable: GDP Growth Rate
b. Predictors: (Constant), Unemployment Rate, Real Interest Rate, Inflation Rate
Coefficientsa
Model
Unstandardized Coefficients
Standardized
t
Sig.
Coefficients
B
(Constant)
Real Interest Rate
Std. Error
27.237
8.350
1.192
.288
.495
-1.280
Beta
3.262
.017
.514
4.144
.006
.426
.188
1.161
.290
.262
-.805
-4.884
.003
1
Inflation Rate
Unemployment Rate
a. Dependent Variable: GDP Growth Rate
The intercept coefficient value was 27.237 with the significance-value of 0.017. This shows that
the intercept was significant and holding interest rate, inflation rate and unemployment constant;
GDP growth rate for South Africa will increase by 27.237.
The real interest rate coefficient value was 1.192 with the significance-value of 0.006. This
shows that interest rate was significant and with 1 percent increase in interest rate, holding
inflation rate and unemployment constant, the GDP growth rate for South Africa will increase by
119.2 percent. Therefore, it can be concluded that, high interest rates raises the cost of borrowing
and thus reduce disposable income, hence causes an appreciation in the exchange rate. The
coefficient does not make any economic meaning.
The inflation rate coefficient value was 0.495 with the significance-value of 0.290. This shows
that inflation rate was not significant and with 1 percent increase in inflation rate, holding
interest rate and unemployment constant, the GDP growth rate for South Africa will increase by
49.5 percent.
The unemployment rate coefficient value was 1.280 with the significance-value of 0.003. This
shows that unemployment rate was significant and with 1 percent increase in unemployment rate,
holding interest rate and inflation constant, the GDP growth rate for South Africa will reduce by
128 percent.
The significance of the regression model was of good fit since the significant-F was less than the
0.05. The coefficient of determination was 0.913, meaning that, 91.3 percent of the data was
captured by the regression line and remaining percentage was not captured by the regression line.
Multiple R showed a positive linear relationship results since the value was 0.956 or 95.6%.
References
https://www.dyingeconomy.com/mundell-fleming-model.html
South Africa
Real Interest Rate
Inflation Rate
Unemployment Rate
GDP Growth Rate
3.279301223
5.017157733
24.63999939
3.168556279
3.882872998
5.723943662
24.72999954
2.396232385
2.509245497
5.776404135
24.55999947
2.485468008
3.56702763
6.136020151
24.88999939
1.413826452
3.667742833
4.509208278
25.14999962
1.321862237
3.278251678
6.594604415
26.54000092
0.664552308
4.647315485
5.181082233
27.04000092
1.157946952
5.893677823
4.504577493
26.90999985
1.487617373
5.38333104
4.124350725
28.46999931
0.113053697
2.313127161
3.223885042
29.21999931
-6.431974826
The data used for the analysis in question 1 and 2 and it was collected from the World Bank page