





































Microsoft Word - ELP-V2N1-p55


Economics, Law and Policy 
ISSN 2576-2060 (Print) ISSN 2576-2052 (Online) 

Vol. 2, No. 1, 2019 
www.scholink.org/ojs/index.php/elp 

55 
 

Original Paper 

Inflation Thresholds, Economic Growth and Investment Planning 

In Uganda 

Rogers Matte1* 
1 Macroeconomic Planning Department, National Planning Authority, Republic of Uganda 
* Rogers Matte, Macroeconomic Planning Department, National Planning Authority, Republic of 

Uganda 

 

Received: October 21, 2018     Accepted: November 7, 2018    Online Published: January 3, 2019 

doi:10.22158/elp.v2n1p55              URL: http://dx.doi.org/10.22158/elp.v2n1p55 

 

Abstract 

Economic Planners, monetary policy custodians and civil society in Uganda often disagree on the 

target for inflation when their development objectives are not harmonised. When development 

economists argue for increased deficit spending in support of infrastructure development and capital 

accumulation, they are challenged in regards how much pressure the development budget should put 

on likely macroeconomic stability, particularly where inflation could rise above the inflation target. 

This paper examined the effect of inflation on economic growth in Uganda and evaluates the 

equilibrium rate of inflation in the country, given the macroeconomic environment. 

Using the threshold model and data for the period 1991-2017 it is established that: a) below 7.3 

percent inflation level, the relationship between inflation and economic growth is positive and inflation 

is not harmful to growth, while at levels above 7.3 percent, inflation was detrimental to economic 

growth and the relationship become negative; b) at economic growth rates above 7.8 percent, inflation 

was an incentive for further growth, yet at economic growth rates below 7.8 percent per annum, 

increases in inflation served as a dis-incentive to economic growth. Therefore Uganda in the current 

conditions is better off maintaining inflation below 7.3 percent as long as the anticipated economic 

growth is 7.8 percent. 

Keywords 

inflation thresholds, economic growth, investment planning 

 

 

 

 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

56 
Published by SCHOLINK INC. 

1. Introduction 

Over the last decade or so Uganda government has been seeking to develop a private led economy 

where private investment and export growth drive economic growth through the national development 

plans and poverty eradication programmes. Attaining a sustained output growth rate capable of 

delivering the country to middle income status in the medium to long-term was the real sector goal of 

the country’s national development plans (Republic of Uganda, 2015). At the same time it has pursued 

a monetary policy goal of macroeconomic stability with the inflation targeting framework in the recent 

years. Price stability is praised for creating a stable environment for decision making by the economic 

agents in markets. High inflation erodes the value of money and affects economic welfare of low 

income earners. At the same time, some moderate inflation is favourable for giving signal to market 

players on supply and demand situation which may affect decision making at agency level. Specific 

plan scenarios assume stable price levels and this makes the issue of inflation thresholds critical to 

economic planning, when the structure of the economy is expected to change over time. 

The debate regarding the management of inflation, its policies and instruments in Uganda overtime 

culminated in setting of a medium to long-term target of 5percent core inflation. The major policy 

framework for inflation in the country since 2011 is inflation targeting. The monetary policy 

framework uses interest rate to signal and influence macroeconomic activity: prices, inflation, output, 

or sometimes designated monetary aggregates. There is no direct control over these and Bank of 

Uganda exerts its influence over the macroeconomic magnitudes by setting the short-term interest rate 

(Mugume & Namanya, 2014). 

Despite adjustments in the central bank interest rate, there is no mention of growth based tools to 

address inflation. In view of the role of the East African Community integration, the country agreed to 

the convergence target of 8 perccent headline and 5 percent core inflation by 2021. The efforts to 

remain around the inflation target in some cases has been criticised for contributing to the high interest 

rates given a high base set by the policy rate. While the relationship between inflation and growth may 

been dogenous, it is important to understand the relationship in the medium to long-term to clearly 

visualise the impacts (Vinayagathasan, 2013).  

1.1 Macroeconomic Stability and Economic Growth 

The primary objective of monetary policy in Uganda over the recent years has been to attain low and 

stable inflation. Close to a decade, the monetary policy stance in Uganda has been guided by policy 

persistence; the neutral nominal rate which takes into account the real interest rate and the expected 

inflation; inflation gap; and output gap (National Planning Authority, 2018). Monetary policy stance 

was contractionary in 2011 and 2015 when inflation was on the rise. The rest of the period saw a 

relaxation in the policy stance so as to increase money growth. When output is at its potential, and 

inflation and inflation expectations are at objective levels, the monetary conditions are at neutral levels 

and do not constrain or stimulate aggregate demand and inflation. In this case real interest rate would 

be at the neutral rate.  



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

57 
Published by SCHOLINK INC. 

Considering the efforts by the central bank to control inflation using interest rates, the growth effect of 

inflation on economic growth is more important in the economics of Uganda as a developing country. 

High lending rates inhibit growth in private investment and the process of value addition, yet these are 

primary growth objectives. This is because the cost and nature of credit has implications on the ability 

of the country to service the entire value chain of products prioritised in the national development plan 

for a country like Uganda. The national development framework expects that the growth of credit 

would be sufficient to support production of raw materials, value addition and marketing and logistics 

management along product value chains. This has however not been realised because the nature of 

credit available in the Ugandan market can to a greater extent serve the interests of the services sector 

and less the processes of value addition and production of primary inputs, which require lower cost 

credit because of long gestation periods. The sectors that are served more in current circumstances are 

more volatile in nature unlike the productive sectors which would provide tangible, stable and 

sustainable growth (Table 1). This process would be more supportive to the transformation process of 

the economy with greater bias towards industrialisation, in particular agro-processing and mineral 

beneficiation.  

In comparison with the countries in the sub-Sahara Africa over the past decade, and more so in the last 

four years, it is observed that interest rates have remained higher in Uganda compared to Kenya, 

Burundi, Nigeria, South Africa, Democratic republic of Congo and Zambia. Compared to these 

countries, interest rates in Uganda have remained around 20 percent since the second half of the 1990s 

unlike other countries where these rates have been moving in response to their economic policies. All 

in all, interest rates and bank credit policies remain relatively weak channels of monetary policy in 

Uganda, unlike the exchange rate policies, in stimulating the real sector performance especially the 

manufacturing sector (Nampewo, Munyambonera, & Mayanja, 2013). Generally, Ugandan interest 

rates have remained above the rest of the countries in Africa both those with a developed banking 

sector and those less developed compared to the domestic sectors. The more developed economies in 

this case include Kenya, South Africa and Nigeria and the less developed ones include Burundi and the 

Democratic republic of Congo. 

Further observed is that the responsiveness of the lending rates in Uganda to changes in the policy 

interest rates of the central bank, shows greater stickiness on the side of lending rates coming down, yet 

a rise attracts immediate response. This supports the view that commercial banks in Uganda are more 

responsive to profit motives than growing the business size, yet they respond positively to policy in a 

direction that is anti-private investment.  

In view of the above observations, the causes of high interest rates in Uganda may be characterised as 

on the basis of internal characteristics of the banks, and the economy wide macroeconomic 

environment. While the lending rates are driven by funding costs, operational costs, capital reserve 

costs, risk cost and bank profits (Bryony, 2012), the macroeconomic environment also presents factors 

that may affect lending rates. The funding cost takes care of the cost faced by banks to raise the 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

58 
Published by SCHOLINK INC. 

necessary capital in order to lend; the operational cost presents the fixed costs faced by banks in terms 

of the overhead running costs; the capital reserve cost is the cost faced by banks in holding the 

minimum capital reserves as required by regulation; the risk cost captures the loss banks suffer from 

default. Macroeconomic determinants of high lending rates include: high and variable inflation, growth 

of output, high money market real interest rates, interest rate uncertainty (inter-bank interest rate 

volatility), exchange rate volatility (Nampewo, 2013), the share of commercial bank public sector loans, 

share of development (long-term) to commercial (short-term) assets, and public sector domestic 

borrowing, among others.  

 

Table 1. Ten Year Share of Total Private Sector Credit in Uganda by Sector 

Sector 
2006/

07 

2007/

08 

2008/

09 

2009/

10 

2010/

11 

2011/

12 

2012/

13 

2013/

14 

2014/

15 

2015/

16 

2016/

17 

Agriculture 6.7 6.0 4.5 6.4 6.7 6.5 7.8 9.6 9.8 10.3 11.5 

o/w 

Production 
2.6 2.3 1.4 3.5 3.6 3.4 3.5 3.8 4.4 4.4 4.5 

o/w 

Processing 

& 

Marketing 

4.0 3.7 3.1 2.9 3.1 3.1 4.3 5.8 5.3 5.9 7.0 

Mining and 

Quarrying 
2.4 0.4 0.3 0.8 0.3 0.4 0.3 0.2 0.5 0.6 0.6 

Crude 

Petroleum 

& Natural 

Gas 

0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.2 0.4 0.3 

Other 

Mining & 

Quarrying 

0.0 0.0 0.0 0.8 0.3 0.3 0.2 0.1 0.2 0.2 0.4 

Manufacturi

ng 
14.1 12.4 15.2 13.2 13.7 13.4 14.0 13.2 15.4 14.0 12.8 

Trade 15.6 12.2 20.6 20.4 22.5 22.6 21.1 21.6 20.4 19.0 20.1 

Transport 

and 

Communica

tion 

6.1 6.9 5.8 7.7 7.7 6.5 5.8 5.4 5.2 7.0 6.7 

Electricity 0.4 0.9 0.6 1.1 0.9 1.0 1.4 1.2 1.7 2.0 1.9 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

59 
Published by SCHOLINK INC. 

and Water 

Building, 

Mortgage, 

Constructio

n and Real 

Estate 

11.2 15.1 16.4 18.2 20.1 22.8 22.6 22.7 22.6 22.8 20.6 

Business 

Services 
0.0 0.0 0.0 3.3 4.3 3.6 5.2 4.4 4.7 3.8 4.1 

o/w 

Working 

Capital 

0.0 0.0 0.0 1.0 1.4 1.7 2.2 1.9 2.3 2.3 2.3 

o/w Other 0.0 0.0 0.0 2.3 2.9 1.9 2.9 2.5 2.4 1.5 1.8 

Community

, Social & 

Other 

Services 

0.0 0.0 0.0 2.9 3.3 3.5 3.1 3.3 3.3 3.4 3.3 

Personal 

Loans and 

Household 

Loans 

17.9 15.4 21.9 21.1 15.8 15.3 13.6 17.2 15.1 15.8 17.7 

Other 

Services 
25.7 30.8 14.6 4.9 4.7 4.3 5.1 1.4 1.4 1.4 0.7 

Total 100 100 100 100 100 100 100 100 100 100 100 

Source: Bank of Uganda. 

 

1.2 Trends and Policy Frameworks for Inflation  

Since 1994, inflation in Uganda was highest in 2011 (quarter 4) at 23.6 percent when the CBR was 

increased to 22 percent. Through this policy there has been a drastic reduction of long term inflation to 

the long term target although there are episodes of high and low inflation in the short run, which are a 

result of shocks whose impact is largely dependent on the structure of the economy.  

The objective of monetary policy in Uganda has over the last decade been that of maintaining 

macroeconomic stability without compromising economic growth. The policy framework in place has 

since 2011 fostered a movement of the interbank money market interest rates in tandem with the central 

bank rate which in turn were expected to influence other retail interest rates in the economy, both in the 

short and the long-term. The target for inflation over the medium to long-term has been 5 percent per 

annum, and the central bank rate has been adjusted accordingly to influence demand for credit and 

influence the level of economic activity by managing the demand side of the economy. Over the 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

60 
Published by SCHOLINK INC. 

medium term, inflation oscillated around the target of 5.0 percent pursued by the central bank, 

especially after the adoption of the inflation targeting framework. When a longer projection is made 

over a period of about twenty years backwards, the expected inflation raises higher taking into account 

the levels attained. It is accordingly noted that the country’s long run optimal level of inflation lies 

between 5 and 8 percent. A target of 6.5 percent has also been estimated from threshold models taking 

into account economic growth in the past and potential output going forward.  

1.3 Sources and Effects of Inflation in Uganda 

Performance of the global and regional economy affects inflation. Uganda has a trade deficit and 

heavily imports manufactured commodities from both regional and global markets. Crises in other 

countries are translated into import prices for the domestic economy. International oil price fluctuations 

have had a key role in influencing domestic prices. Volatility in the foreign exchange market is another 

source, especially depreciation fuels inflation. Uganda currency has depreciated significantly in the 

recent past leading to secondary effects on the domestically produced commodities given a large 

component of imported raw materials. 

Large budget deficits induce increases in interest rates and high cost of private capital, with effects on 

inflation. There has been an increase in the domestically financed borrowing which leads to the rise in 

the cost of credit in the banking system, impacting on the cost of access to capital and of production. 

Delays in execution of public projects for which resources have been borrowed domestically 

exacerbate this issue and may have led to sub-optimal decisions. 

Uganda has underdeveloped commodity value chains and markets which provide an incentive for 

supply rigidities. The dependence of the food sector on natural factors and the lack of linkages with 

industry and under developed market institutions affect the variability in food prices in periods of boom 

and shortage. Further the under developed utilities sector affects production to capacity for some 

industries therefore affecting the cost of production.  

Food and Non-Alcoholic Beverages; Housing, Water, Electricity, Gas and Other Fuels; Transport and 

education were the major sources of inflation in the country from 2011 to 2017, contributing over 

63percent of the price changes. The prices for food and non-alcoholic beverages are associated with the 

structural nature of agriculture production in the country, with a large percentage resulting from the 

shortages in supply. Addressing the key real sector factors inhibiting stable supply of food, constraints 

in the supply of fuel products in the transport sector and educational supplies could potentially reduce 

inflation from these sources. 

The description of inflation dynamics in Uganda during the period 2000-2012 indicated that in the 

long-run money supply, exchange rate, foreign inflation, terms of trade and real output, foreign prices, 

exchange rate, growth of domestic credit, rainfall deviation from the long-run mean, the trade and 

current account balances, fiscal balance, trade openness, and the international interest rate differential 

determine inflation in Uganda (Opolot & Kyeyune, 2012; Janine, Muellbauer, & Sebudde, 2015). In 

the short run, inflation is driven by changes in real output, monetary aggregates, the exchange rate, and 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

61 
Published by SCHOLINK INC. 

foreign prices. The Central Bank of Uganda indicates that core inflation is affected by: international 

factors (interest rates, inflation, oil prices); exchange rates; output gap and capacity utilization; 

domestic demand; money supply and credit extension and expectations (Bank of Uganda, 2011). The 

disequilibria in the money and traded goods markets are significant but the adjustment process is slow. 

In spite of the fact that inflation has exogenous determinants, such as foreign inflation, there is scope 

for the central bank to limit the impact of such shocks on inflation by pursuing a tight monetary policy 

stance. The significance of the interest rate differential in the money-demand equation implies some 

degree of effectiveness of the monetary policy. 

1.4 Effects and the Threshold of Inflation  

Albeit many arguments regarding inflation, there is a view that inflation is useful to economic growth if 

below some threshold level, and is detrimental when above that threshold level. In particular, inflation 

influences growth by decreasing productivity growth and investment. Uganda’s implied inflation 

threshold is 5 percent as set in the targeting framework by the central bank—Bank of Uganda, which is 

the point of variance. This threshold is unique for each economy, as indicated by the variations seen in 

cross-country studies, between 2 percent and 12 percent in a number of countries, and generally, 

developed countries have lower thresholds while those for developing countries are higher. A recent 

study estimated 8.5 percent as the threshold for Uganda based on data for the period 1970-2013 (Yabu 

& Kessy, 2015), yet the inflation target by BoU is 5 percent. What is not clear is whether this threshold 

is dynamic. The EAC countries under the programming for the anticipated Monetary Union target to 

have headline inflation at 8 percent. Whether this target would be changed and how often is not clear 

either.  

In view of the above, development planners and monetary policy custodians disagree on the target for 

inflation when their development objectives are not harmonised. Whereas development economists 

may argue for increased deficit spending in support of capital accumulation, they face a challenge in 

regards how much pressure the development budget should put on likely macroeconomic stability 

especially where inflation would rise above the inflation target.  

1.5 Theoretical Review 

A number of theories explain the link between inflation and economic growth. There is confluence 

among them that in the short run, inflation induces growth, to the extent that it is positively correlated 

with growth but in the long run its persistence is detrimental to growth. Table 2 reviews some of the 

theoretical impacts of inflation. 

 

Table 2. Theoretical Economic Growth Impacts of Inflation 

Theory Nature Inflation and growth Criticism Emphasis 

Classical Supply side 

based 

A rise in inflation leads to a fall 

in the rate of return on 

High inflation 

inhibits financial 

Provide 

incentives for 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

62 
Published by SCHOLINK INC. 

Theory Nature Inflation and growth Criticism Emphasis 

individual’s real money balances 

reducing wealth. People will 

then save more by switching to 

financial assets in order to 

accumulate the desired wealth. 

Higher demand for assets 

increases their prices, resulting 

in a decline in the real interest 

rate. Increased savings result in 

greater capital accumulation and 

hence faster output growth 

development by 

making financial 

intermediation 

more costly, hits 

the poor 

disproportionately 

because they do 

not hold financial 

assets that 

provide a hedge 

against it. This 

hinders long-term 

economic growth 

increased 

savings and 

investment so 

that the 

economy 

grows 

Keynesian AD-AS   “ “ 

 Inflation can redistribute 

profits from workers with low 

savings propensities to 

entrepreneurs with high 

propensities to save and invest, 

and increases the nominal rates 

of return relative to the cost; 

hence increasing growth 

 During inflationary periods, 

money is redistributed from 

holders of money balances to 

governments (through monetary 

authorities – inflation tax) to 

expand public investment 

programmes, thus increasing 

growth 

 

 

High inflation 

rates raise the 

cost and risk of 

productive capital 

and may lead to 

misallocation of 

funds to less 

productive 

investments that 

act as a hedge 

against inflation 

 

… 

Neoclassical 

and 

endogenous 

growth 

inflation affects 

capital 

accumulation 

and investment; 

“ “ 

 

High inflation 

leads to 

speculative trade 

and capital 

… 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

63 
Published by SCHOLINK INC. 

Theory Nature Inflation and growth Criticism Emphasis 

outflows affecting 

growth 

Monetarist the role played 

by monetary 

growth in 

determining 

inflation – from 

the quantity 

theory of money 

There is neutrality of money, 

with inflation rate having no 

effect on the growth rate as well 

as the level of output. In the 

long run, the growth rate in 

money mainly affects prices. 

This is because individuals will 

anticipate the rate of future 

inflation and incorporate its 

effects into their behaviour. 

Employment and output are not 

affected 

 Provide 

dis-incentive 

for monetary 

growth to 

contain 

inflation. 

When profits 

decline and 

labour 

productivity 

remain 

unchanged, the 

prices tend to 

increase 

 

In view of the above theoretical frameworks, a number of models have been used to analyse the links 

between inflation and economic growth. These include the Philips curve, the mark-up approach, the 

quantity theory of money, and the dynamic panel. These are briefly reviewed below. 

Phillips Curve: If demand increases and the output gap becomes larger, business costs will increase, 

wages will increase. Companies pass on the higher costs to their customers, thus increasing prices. On 

the other hand, if demand falls and the output gap becomes smaller, business costs will decline thus 

wages will be lower. Companies then pass on the lower costs to their customers by reducing prices, 

hence less inflation. The Philips curve therefore underscores the link between the output gap and the 

inflation rate, with current inflation explained by current inflation expectations and the output gap. 

1 1

e
t t

t t
t t

dP dP
EP

P P
   

 

                               (1) 

In the equation above, the LHS is the inflation rate (
1

t

t

dP

P

), the current expectation of the future 

inflation is the second term on the RHS (
1

e
t

t

dP

P

), while the output gap is the third term ( tEP ). The 

signs of the coefficients are all positive in the equation above.  

The Mark-up approach: Considering Unit Labour Cost (ULC), Price of imported goods (PM), then 

enterprises will set domestic prices by adding a gross mark-up () to the cost per unit of output. Here 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

64 
Published by SCHOLINK INC. 

the CPI combines the price of domestic and imported goods.   

log(Pt+1)=+*log(C*U*Mt+1)+*log(PMt+1)+t+1                 (2) 

The quantity theory of money: Considering Money supplies (M), Velocity of money (V), Real GDP 

(Y), then money growth is a key determinant of inflation. By assuming that Money demand remains 

stable, then we estimate current prices as: 

1 1

1

1

(1 ) * (1 )

1 (1 )
(1 )

t t

t t t
t

tt

t

dMd dV

dP Md V
dYP
Y

  





 
   


                      (3) 

The threshold model: In this model, the growth rate of real GDP is dependent on inflation, the inflation 

threshold variable that is exogenous and a time variant, and other explanatory factors. These 

explanatory factors include the past values of GDP, investment, population growth, openness, terms of 

trade, and the standard deviations of openness and terms of trade. The model is divided into two using a 

dummy variable for values of inflation below and above the threshold.  

This is a form of nonlinear regression presenting piecewise linear specifications and regime switching 

that occurs when an observed variable crosses unknown thresholds (IHS Global Inc, 2016). In this 

model, there are T observations and m potential thresholds (j=1, 2, … m), implying there are (m+1) 

regimes. For all observations in the regimes, there are two sets of regressors: those that vary (X) and 

those that do not vary (Z) with the regime. For the threshold variable (qt) threshold values (m) exist, 

such that we can identify them across the arrange j<=q<j+1, and (1<2 < …. <m). For a single 

threshol and two regimes, we have yt=Xt’+Zt’1+et,if–<qt<1; and yt=Xt’+Zt’2+et,if–1<=qt<. 

For multiple regime specifications, a single equation can be used to combine them. For 1(.) an indictor 

function for which the expression is true and 1 ( , )j qt 
 

carries the value 1 and zero otherwise.  

0

1 ( , ) * '
m

t j t j t
j

yt X qt Z   


   ;                   (4) 

This is a nonlinear least squares estimation approach, in which there is need to find the coefficients  

and ; the threshold values ; and identify the threshold variable qt using model selection. The 

objective threshold function, S(,,), is then minimized with respect to the parameters. 

2

1 0

( , , ) 1 ( , ) * 't

T m

t j t j
t j

S y X qt Z     
 

 
   

 
                 (5) 

Eviews provides a userfriendly interface for threshold modelling taking into account the above 

framework (IHS Global Inc, 2016). This apprach is used in comparison with the specification of the 

function in a loglinear form for the threshold ariables and manually estimating the threshold by solving 

the derivative of the function with respect to the threshold variable. 

 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

65 
Published by SCHOLINK INC. 

1.6 Major Cross Country and Historical Experience 

The study by Adegoke (2012) highlights the key studies that have been conducted on this subject, in 

which for instance the existence of threshold effects in the inflation-growth relationship of Nigeria, 

from 1970 to 2003 was established at the level of six per cent. It is albeit noted that the threshold 

changed as the structure of the economy evolved to eight percent (Adegoke, 2012). Table 3 reviews 

some of the causes of inflation in selected case studies. 

 

Table 3. Selected Determinants of Inflation Cross Countries 

Research Country Causes of Inflation Dominant factor 

Atta et al. (1999)  Botswana nominal money supply, 

nominal interest rate, nominal 

exchange rate, nominal wages 

and South African consumer 

price index, US exchange rate 

pass-through 

monetary variables 

(money supply and 

interest rates) 

Dlamini et al. (2001), 

Akinboade et al (2004) 

Swaziland and 

South Africa 

real sector effects 

(GDP) 

Ocran et al (2005) Ghana  inflation persistence 

Khan et al. (2007), 

Leheyda (2005) 

Pakistan and 

Ukraine 

money supply, wages and 

exchange rates, fiscal policy 

price inertia 

Source: Haile (n.d.). 

 

In the 1960s there was a view that inflation was positively correlated with economic growth in the short 

run, and to some degree, in the long run. Consensus in the 1970s and 1980s was that of a positive short 

run relationship between growth and inflation, such that while stabilisation of hyperinflation had little 

output costs, sterilisation of mere high inflation was on the otherhand costly. Most recent research since 

the 1990’s was concerned with the longrun relationship and advanced an inverse relationship between 

inflation and growth.  

While investigating the existence of a threshold level for inflation and how any such level affected the 

growth, a dynamic panel threshold growth regression was used for 32 Asian countries over the period 

1980-2009, a threshold of approximately 5.4 percent was estimated. While inflation was found to hurt 

growth when it exceeded 5.4 percent it had no effect below this level (Vinayagathasan, 2013). 

1.7 Investment Planning, Fiscal Deficits and Inflation  

From Uganda’s experience, during planning processes planners are concerned about the role of fiscal 

deficits and external borrowing and the effect of foreign exchange inflows on inflation. The value of 

invesments needs to keep in tandem with the levels of inflation. In the NDPII planning process, it was 

argued that addressing the effects of inflationary expenditure, given high deficit financing would have a 

slowdown effect on private sector credit when interest rates are raised. As a consequence, this would 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

66 
Published by SCHOLINK INC. 

slow down economic growth below desided levels (National Planning Authority, 2015). Further, the 

planners argued that allowing inflation to increase would have an effect on the depreciation of the 

shilling, yet the central bank has a limit on issuing liquidity to address the effects of public 

administration expenditures.  

1.8 Research Questions 

The main research question that this paper seeks to answer is: what is the optimal level of inflation for 

Uganda above which the development budget may not result into lower economic growth. Secondly, 

with the current structure of the economy, what is the appropriate level of growth resulting from 

increased investment that may not lead to higher inflation? 

1.9 Research Objectives 

This paper therefore examines the effect (positive or negative) of inflation on economic growth in 

Uganda; and evaluates the equilibrium rate of inflation in the country, given the macroeconomic 

environment. 

 

2. Method 

This paper adopts the threshold model as illustrated by Yubu and Kessyto estimate the threshold for 

EAC (Yabu & Kessy, 2015) in a quadratic specification and compares the results with those generated 

by the threshold regression model as specified in IHS Global Inc (2016). Unlike the method used by 

Adegoke (2012), the threshold model is more elaborate and does not involve arbitrary selection of the 

threshold levels. However, we differ from Yubu and Kessy by estimating the ADL rather than the use 

of error correction and cointegration.  

2.1 Model Specification 

In this model, the threshold level of inflation is obtained is obtained in a combination of linear and 

squared term for inflation such that the impact of inflation on economic growth with positive effects of 

inflation switching to negative when inflation exceeds some threshold level. At the threshold level of 

inflation, the function is at maximum after which the marginal effect of inflation becomes negative. 

DLOG(RGDP)*=1*INF_RATE+2*INF_RATE(-1)+3*(INF_RATE)2+4*DLOG(POPTOT)+ 

5*DLOG(POPTOT(-1))+6*DLOG(POPTOT(-3))+7*PSC_GR+8*PSC_GR(-1)+ 

9*DLOG(INVTit)+10*DLOG(INVT(-1))+11*DLOG(GER)+12*RER_GR(-1)+ 

13*DLOG(TOT_2005)+14*DUMMY1+Mu                    (6) 

Where DLOG(RGDP)* is growth rate of real GDP, INF_RATE is growth rate of CPI and DLOG 

(POPTOT) is population growth rate, PSC_GR is private sector credit growth, DLOG (INVT) is 

investment growth, DLOG (GER) is growth in gross external reserves, DLOG (TOT_2005) is 

percentage change in terms of trade at 2005 prices for openness, and DUM is a dummy variable, which 

takes zero during inflation targeting and one elsewhere; and Mu is the error term. The peak of the 

function identifies the critical point of inflation above which the marginal impact of inflation on growth 

is negative and is calculated as the inflation threshold level. This is established by finding the partial 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

67 
Published by SCHOLINK INC. 

derivative of the function of Growth with respect to inflation and setting it to zero. As observed in the 

Table 5 some of the variables are excluded from the OLS method to estimate manually equation 9, 

while others are introduced in the threshold method.   

dlog(RGDP)/dlog(INF_RATE)=1+23*INF_RATE=0           (7) 

This implies that the critical level of inflation is the ratio: -1/23.  

2.2 Data  

The time series data used for this analysis is for the period 1990 to 2017, and the variable definitions 

and their data sources are described in the Table 4.  

 

Table 4. Description of Data and Its Sources 

Variable Data Collected Sector and Source 

Growth rate of real GDP  Real Gross Domestic 

Product (GDP) 

National Accounts data, Uganda Bureau of 

Statistics (UBOS) 

Population growth rate  Population Size Population Statistics, UBOS 

Investment  Gross Fixed Capital 

Formation 

National Accounts Data, UBOS 

Terms of Trade (TOT) Terms of trade National Accounts data, UBOS 

Growth rate of CPI Consumer Price Index 

(CPI) 

Real Sector Statistics, Bank of Uganda (BoU) 

Credit to GDP ratio  Credit to the Private 

sector 

Monetary and Financial Sector Statistics, 

Bank of Uganda, BoU 

Gross External Reserves Gross External 

Reserves 

External Sector Statistics, Balance of 

Payments according to 6th Edition of the 

Manual, BoU 

Real Exchange Rates Real Exchange Rate 

Index 

External Sector Statistics, Balance of 

Payments according to 6th Edition of the 

Manual, BoU 

 

2.3 Model and Data Diagnostics  

Using EViews in built tests for normality, serial correlation, and heteroskedasticity undertaken. In 

particular, we estimate the Jarque-Bera and probability values for normality; the Breusch-Godfrey 

statistics for serial correlation, and the Breusch-Pegan-Godfrey statistics for heteroskedasticity. The 

procedure in EViews Jarque-Bera and probability values is that after estimating the equation, we 

view/seek residual tests, and then select white heteroskedasticity. Obtained are the test statistic and the 

probability so that the statistic exceeds (in absolute value) the observed value under the null hypothesis. 

We reject the null hypothesis that the residuals are normally distributed if the value of the probability is 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

68 
Published by SCHOLINK INC. 

small, otherwise accept. 

The heteroskedasticity demonstrates that the there is no auto regressive conditional heteroskedasticity 

in the residuals. The procedure finds out whether the series are not related to the recent residuals, to 

safeguard the efficiency of the model. The hypothesis that there is no autoregressive conditional 

heteroskedasticity in the residuals is tested, and the test statistic is estimated from the regression of the 

squared residuals on a constant and squares of its lagged values up to an appropriate order. 

 

3. Results 

This section presents results of the Inflation—Growth trade-offs and its implications for investment 

planning. The threshold levels of inflation have been obtained in a linearized function and the impact of 

inflation on economic growth established to show positive effects of inflation switching to negative 

when inflation exceeds the observed threshold level. Table 5 presents the results of the regression, and 

the models fulfil the conditions for Normality, Serial Correlation, and Heteroskedasticity Test. 

The behaviour of changing marginal effect of inflation is observed and becomes negative after the 

threshold. Model 1 Used the least squares estimation, equation 5 is estimated from equation 4 to 

generate data for the threshold. 

The threshold variable in the OLS method was the rate of inflation (INF_RATE) while in the threshold 

method, both inflation rate and the percentage economic growth rate (DLOG (RGDP)*100) were used. 

Both methods present the plausible assigns and statistical significance of the coefficients as indicated in 

Table 5. While in the OLS method the estimation of the threshold variable is straight forward, in the 

threshold regression a choice is made based on the significance of the model coefficients. In the 

threshold model, the choice of threshold variable was the economic growth rate, and the value used was 

7.9 percent. In the OLS method, the value estimates using equation 5 is 7.3 percent. 

The results from the OLS Method imply that, below 7.3 percent inflation level, the relationship 

between inflation and economic growth is positive and inflation is not harmful to growth. At levels 

above 7.3 percent, inflation is detrimental to economic growth and the relationship becomes negative. 

On the other hand, the threshold method identified two regimes, one in which the economic growth rate 

is above 7.85 percent, and another where economic growth rate is equal to or less than 7.85 percent. 

 

Table 5. Results of the Inflation, Growth Thresholds 

Variable 
Coefficients 

Least Squares Method Threshold Regression Method 

Threshold Choice 
INF_RATE=-7.3=0.005384/(2*-0.00

0367) 

DLOG(RGDP)*100< 

7.853817 (17 obs) 

7.853817 

<=DLOG(RGDP)*100(8 

obs) 

INF_RATE 0.005384**[0.002037](2.643113) -0.165432***[0.033270] 0.380128***[0.059707](6.3



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

69 
Published by SCHOLINK INC. 

(-4.972368) 66513) 

INF_RATE(-1) 0.000359[0.000582](0.617014)  

INF_RATE^2 -0.000367***[0.000106](-3.45947

9) 

 

DLOG(POPTOT) 8.530238***[2.314828](3.685041) -193.8198[148.2772] (-1.307145) 

DLOG(POPTOT(-1))  349.1653*[150.1007] (2.326207) 

DLOG(POPTOT(-3)) -8.012950***[2.322137] 

(-3.450679) 

 

PSC_GR 0.000438*[0.000236] (1.858553)  

PSC_GR(-1)  0.048367***[0.007903] (6.120030) 

DLOG(INVT) 0.234771***[0.050379](4.660058) 17.76013***[3.974106](4.468962) 

DLOG(INVT(-1))  -9.018084**[3.162211](-2.851828) 

DLOG(GER) -0.051089*[0.024192](-2.111787) -1.059403*[0.601698](-1.760688) 

RER_GR(-1)  -0.059789*[0.034311](-1.742554) 

DLOG(TOT_2005)  -1.625303[1.588881](-1.022922) 

DUMMY1 0.025210**[0.008824](2.857144) 0.419792[0.454022](0.924608) 

R2 0.82 0.95 

Adj R2 0.72 0.92 

D-W stat 2.55 2.34 

Other Notes Threshold variable considered 

(manual): INF_RATE; Sample 

(adjusted): 1995 2017;  

Included observations: 23 after 

adjustments; white 

heteroskedasticity-consistent 

standard errors & covariance 

Sample (adjusted): 1993 2017; Threshold type: 

Bai-Perron tests of L+1 vs. L sequentially 

determinedthresholds; Threshold variables 

considered: INF_RATE, LOG(RGDP)*100; 

Threshold variable chosen: DLOG(RGDP)*100; 

Threshold selection: Trimming 0.15, Max. 

thresholds 5, Sig. level 0.05; Threshold value used: 

7.853817; White heteroskedasticity-consistent 

standard errors &covariances 

Model Diagnostics   

Normality JarqueBer

a 

1.7034

14 

Prob 0.426686 JarqueBera   

1.013846 

Prob:                         

0.6023 

Breusch-Godfrey Serial 

Correlation LM Test 

F-stat  1.344

756 

Prob. 

F(9,13) 

0.2972 F-statistic    

0.352411 

Prob. F(2,12)                

0.7100 

Obs*R2 4.211

084 

Prob. 

�2(2) 

0.1218 Obs* R2        

1.386919 

Prob. �2(2)    0.4998 

Breusch-Pagan-Godfrey F-stat  0.760 Prob. 0.6533 F-statistic    Prob. F(11,13)           



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

70 
Published by SCHOLINK INC. 

Note. SE [ ]; t-Statistic ( ); Level of significance: “*”=10 percent; “**”=5 percent; “***”=1 percent. 

 

4. Discussion  

At very high economic growth rates above 7.8 percent, inflation is an incentive for further growth. 

However, when growth is below 7.8 percent per annum, increases in inflation serves as a dis-incentive 

to economic growth. Therefore Uganda in the current conditions is better off maintaining inflation 

below 7.3 percent as long as the anticipated economic growth is at least 7.8 percent. In comparison 

with the existing research, this is consistent with the EAC macroeconomic convergence criterion which 

recommends inflation below 8 percent. As observed earlier in the literature, the study on the founding 

members of the EAC estimated a threshold of 8.4 percent but was silent on the economic growth 

threshold in this relationship (Yabu & Kessy, 2015). Further, these findings are consistent with other 

research conclusion that the threshold at which inflation reduces growth is in the single-digits (Heintz 

& Ndikumana, 2010; Ghosh & Phillips, 1998; Burdekin, Denzau, Kei, Sitthiyot, & Willett, 2004). 

In this framework, the non-threshold factors that significantly affect the threshold relationship include: 

population growth rate, the total investment, and the accumulation of gross external reserves from the 

external sector. A dummy variable introduced to identify two economic regimes of inflation targeting 

and that before inflation targeting is significant in the OLS model unlike the threshold method. In the 

OLS it signifies the change in policy on inflation management. It would be important to assess the 

significance for rebasing regime for the economy after 2009.  

Population growth while having significant effect on economic growth rate, its marginal impact has 

mixed results. Population growth in itself may not be important unless it contributes the needed quality 

labour force that participates effectively in the value chains of economic products. Growing purchasing 

power per capita coupled with human development is important for sustained economic growth. 

Lagged population growth could signify the accumulation of human capital in the growth—inflation 

trade-off as a result of innovations.  

Private sector credit induced economic growth in both methodologies. Monetary policy has used 

changes in credit to influence inflation and create stability in the growth inducing environment. Growth 

in total current investment is primarily a critical factor for economic growth due to its contribution to 

capital stock. The accumulation of gross external reserves though significant was a dis-incentive to 

economic growth, though openness was positive but highly significant. Similarly, current changes in 

terms of trade were a dis-incentive to economic growth though not significant.  

In conclusion, this study has considered the inflation-development nexus in Uganda using data for the 

period 1991 to 2017. The study methods used estimate threshold variable in a threshold regression. 

Using the threshold variables of the rates of inflation and economic growth it is found that, below 7.3 

Heteroskedasticity Test 613 F(9,13) 1.411915 0.2740 

Obs*R2 7.933

624 

Prob. 

�2(9) 

0.5408 Obs*R2   

13.60891 

Prob. �2(11)                

0.2554 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

71 
Published by SCHOLINK INC. 

percent inflation level, the relationship between inflation and economic growth is positive and inflation 

is not harmful to growth. Above 7.3 percent, inflation is detrimental to economic growth and the 

relationship becomes negative. On the other hand, at very high economic growth rates above 7.8 

percent, inflation is an incentive for further growth, but at economic growth rates below 7.8 percent per 

annum, increases in inflation serve as a dis-incentive to economic growth. Therefore Uganda in the 

current conditions is better off maintaining inflation below 7.3 percent as long as the anticipated 

economic growth remains below 7.8 percent.   

The population growth rate, the total investment, the accumulation of gross external reserves and the 

economic policy regimes are important for Uganda’s growth—inflation nexus. Supply side factors are 

important in this framework. Therefore, anti-inflationary strategies in the country should take into 

account the elimination of supply constraints, increasing competition, and increasing capacity and 

efficiency of investment outputs.  

 

References 

Bryony, W. (2012). A Decomposition of Ugandan Lending Rates. Bank of Uganda Working Paper 

Series No. WP/BoU/2012/07. 

Burdekin, R. C., Denzau, A. T., Kei, M. W., Sitthiyot, T., & Willett, T. D. (2004). When does inflation 

hurt economic growth? Different nonlinearities for different economies. Journal of 

Macroeconomics, 3, 519-532. https://doi.org/10.1016/j.jmacro.2003.03.005 

Ghosh, A., & Phillips, S. (1998). Warning: Inflation may be harmful to your growth. International 

Monetary Fund Staff Papers, 4, 672-686. https://doi.org/10.2307/3867589 

Haile, K. (n.d.). The Determinants of Inflation in Botswana and Bank of Botswana’s Medium-Term 

Objective Range. Botswana Journal of Economics, 67-74. 

Heintz, J., & Ndikumana, L. (2010). Is there a case for formal inflation targeting in sub-Saharan Africa? 

In African Development Bank Working Paper Series, Working Paper No. 108 (pp. 1-45). 

IHS Global Inc. (2016). EViews 9 User’s Guide II. Irvine CA: IHS Global Inc. 

Janine, A., Muellbauer, J., & Sebudde, R. (2015). Inflation forecasting models for Uganda: Is mobile 

money relevant? CSAE Working Paper WPS/2015-17, 1-66. 

Mugume, A., & Namanya, C. R. (2014). The Augmented Taylor Rule and setting of Monetary Policy 

Rates in a Developing and Small Open Economies: Bank of Uganda’s Experience. In Bank of 

Uganda Working Paper Series, Working Paper No. 08/2014 (pp. 1-37). 

Nampewo, D. (2013). What Drives Interest Rate Spreads in Uganda’s Banking Sector? International 

Journal of Economics and Finance, 5(1), 76-85. 

Nampewo, D., Munyambonera, E., & Mayanja, L. (2013). Sectoral effects of monetary policy in 

Uganda. Journal of Statistical and Econometric Methods, 2(4), 17-40. 

National Planning Authority. (2015). Report on the Tripatite Discussions on the Scenarios of the 

Macroeconomic Frmework for NDPII, 12-19 January 2014. Kampala: National Planning 



www.scholink.org/ojs/index.php/elp                   Economics, Law and Policy                        Vol. 2, No. 1, 2019 

72 
Published by SCHOLINK INC. 

Authority. 

National Planning Authority. (2018). 10 Year Perspective Macroeconomic Situation Analysis (draft 

Report). Kampala: NPA. 

Opolot, J., & Kyeyune, A. M. (2012). Infllation Dynamics in Uganda: The role of disequilibria in the 

money and traded goods markets. Bank of Uganda Working Paper No. 02/2012. 

Republic of Uganda. (2015). The Second National Development Plan 2015/16-2019/20. Kampala: 

National Planning Authority. 

Vinayagathasan, T. (2013). Inflation and Economic Growth: A Dynamic Panel Threshold Analysis for 

Asian Economies. Tokyo: National Graduate Institute for Policy Studies. 

Yabu, N., & Kessy, N. J. (2015). Appropriate Threshold Level of Inflation for Economic Growth: 

Evidence from the Three Founding EAC Countries. Applied Economics and Finance, 127-144. 

https://doi.org/10.11114/aef.v2i3.966 

Bank of Uganda. (2011, July 25). Understanding the Inflation Targeting—Lite Policy Framework. 

Bank of Uganda. Retrieved October 20, 2018, from https://www.bou.or.ug 

Adegoke, I. A. (2012, June). Inflation and Economic Growth: An Estimate of the Threshold level of 

Inflation in Nigeria. AJEP, 19(1), 81-106. 

 


