Journal of Management Vol. 14, Issue. 2, 2019 ISSN: 1391-8230 1-12 THE IMPACT OF UNEMPLOYMENT AND INTEREST RATE ON INFLATION IN

S. Selvanayagama, A. M. M. Mustafab

a Bank of Ceylon, Fauzi Mawatha, Kattankudy-02. Sri Lanka.

b Faculty of Management and Commerce, South Eastern University of Sri Lanka, Oluvil # 32360. Sri Lanka.

Abstract Three major economic indicators such as Inflation, unemployment and interest rate have an important role in an economy in terms of sustainable development. The long-term progress of the Sri Lankan economy is destabilized. The linkage or the impact among these variables is very important for developing country such as Sri Lanka to overcome the destabilized hurdles. The study intends to investigate the impact of unemployment and interest rate on inflation in Sri Lanka. Also, this study was analyzed the short and long run relationship among the variables. Phillip’s relationship between the variables inflation and unemployment also was discussed in details. Fifty-three years of annual data for period of 1953- 2015 of the variables inflation, unemployment, interest rate, money supply (M2) and government expenditure used for the analysis. Parametric and non-parametric approaches have been employed in this study. The Autoregressive Distributed Lag (ARDL) model with co-integration technique has been employed to find the short and long run relationship of the variable. The statistical package EViews 9 and Microsoft excel were used for the analysis. The study reveals that unemployment is negatively impact on inflation in short and long run in Sri Lanka, which is statistically significance. Further, the study revealed that the Phillip’s relationship between inflation and unemployment exist in Sri Lankan economy. The interest rate is also negatively impact on inflation in short run and positively impact in long run. Results are statistically significance at 5% confidence level and theoretically expected. This study recommends that the relationship between the variables should be noted and utilized the Engine of growth concept in order to achieve sustainable development of Sri Lanka. Job opportunities to be extended further more. Further, the study suggests that using quarterly data to analysis this kind of time series will reflect relationship accurate.

Keywords: Inflation, Unemployment, Interest rate, Money supply, Government Expenditure

Introduction enough supply of goods and services than The unemployment, interest rate and inflation demand exists; which reduces the price level play a vital role in sustainable economic or inflation. The rising purchasing power of growth of any country. The most fundamental people (PPP) leads savings and investment objective of macroeconomics is to sustain which support the economy in an upward high economic growth through maintenance trend of growth rate. Sri Lanka has remained of low rate of unemployment and low rate of as a developing country since a long time and inflation (Adeyi Emmanuel ola, 2012). A the life cycle of the economic development country may fulfill this kind of situation still at an early stage. The economies of where there is full employment of the labour countries like “04 tigers of Asia” such as force operates in an economy. Full Singapore, Taiwan, Hong Kong and South employment means that, no or low Korea are growing faster than Sri Lanka. unemployment rate in an economy creates These four countries positioned as “04 Tigers

[email protected] https://orcid.org/0000-0001-5836-0058 of Asia”, emerging economies and some important findings reviewed as challenging economies to the developed follows.The scholar Handapangoda (2005) nations such as the United States of America examines the relationship between inflation (USA) and United Kingdom (UK). In 1960, and unemployment in Sri Lankan context. the Gross Domestic Product (GDP) of Sri The study concentrates on two different Lanka and Singapore was United States policy regimes such as inward looking in the Dollar (USD) 1.42 billion and USD 0.7 period of 1970 to 1976 and outward looking billion respectively. But, at present GDP of for the period of 1977 to 2005. The study Sri Lanka and Singapore is USD 78.82 found that there is no any significant Billion and USD 307.97 Billion respectively relationship between unemployment and as at 2014. Even though, this kind of inflation in Sri Lanka. That means the real backwardness is unbelievable, it is the real relationship explained by Philip’s does not scenario in Sri Lanka. The Sri Lankan exist in Sri Lanka. Lalani (2014) studied economy grows in addition term, while other about the relationship of inflation and countries grow in multiple terms. According unemployment in Sri Lanka. The author to the Gross National Income (GNI) per analysed the data for the period of 1980-2013 capita of 2014, Singapore and Hong Kong by using Ordinary Leased Square (OLS) and ranked above the UK and USA (World Bank, concluded that there is a positive relationship 2015). Sri Lanka is far behind even though between inflation and unemployment in Sri having almost all the resources, including Lanka. Niranjala (2014) studied the nature of natural and human capital than Singapore. the relationship between unemployment and The impact of 30 years of civil war, which inflation in Sri Lanka. A sample of 35 years resulted in high security expenses, also for the rate of inflation and unemployment has negatively affected the economy. But it is not been taken from the year 1977 to 2012. The ADF only the reason for the failure of the economy. test was used for finding the unit root or stationary The relationship between inflation and and the Granger causality test and OLS was unemployment and related policy employed to find the relationship between the implications also accountable for the failure inflation and unemployment in Sri Lanka. of economy. Hence, finding the relationship The results of the statistical analysis show among these variables in the short and long that there is no trade-off between inflation run is very important in order to understand and unemployment in Sri Lanka. Ewing and the real economic scenario in Sri Lanka. Seyfried (2004) examined the Phillips curve Therefore, it is essential to find the by using the price adjustment mechanism and relationship between inflation and the conditional variance of inflation also unemployment in Sri Lanka whether the included rather than the traditional model, relationship explained by Phillips or Lucas is they suggested that the adding variability of exist in Sri Lanka in order to make effective inflation gives more weight to the model than policy decisions. Hence, the study tries to traditional one. understand the relationship between unemployment, interest rate on inflation in Chaido, and Melina (2012) studied inflation, Sri Lanka. unemployment and the NAIRU in Greece. This paper examines the Phillips Curve Review of literature approach in Greece using annual data from 1980 - 2010. The results show that there is a Reviews the scholars’ view on the particular long run and a causal relationship between field of study related to inflation, inflation and unemployment for the period. unemployment and interest rate. This field of Hussein Ali Al-Zeaud, (2014) investigated study extensively researched all over the the existence of a trade-off relationship world. However, there are only a few scholars between unemployment and inflation in the researched on this field in Sri Lanka. Here, Jordanian economy between 1984 - 2011.

2

Alfred & King (2012) examined the 1996-2012 using the ARDL model and relationship between inflation and revealed that significant and negative effect unemployment in the long run, using of inflation and unemployment on economic quarterly US data from 1952 - 2010. growth in long term. Based on the, review of According to the band-pass filter approach, literature a research question was developed the study found strong evidence that a in this study. That is; What kind of short and positive relationship exists in short and long long run relationship exist among the run. Gulistan, et.al. (2015) studied the variables unemployment, interest rate and relationship between unemployment and inflation in Sri Lanka? inflation in Turkey. This study period is the years between 1988 - 2002. Vector Objective of the study Autoregressive Model (VAR) and impulse- The primary objective of this study is to response analysis are used in the study to investigate the impact of unemployment and explain the relationship. This result is a interest rate on inflation in Sri Lanka. Other supportive of Philips curve. Also, suggest that sub objectives include; the inflation reduces the purchasing power of i. To examine the Phillip’s relationship of the consumers and increase the social issues variables, inflation and unemployment. and creates the social unrest in the country. ii. To test the stationary properties of the Hence, maintaining inflation and time series data. unemployment at optimum level is better to iii. To examine the long run relationship the economical politics. Anthony, among the variables, inflation, et.al.(2015) examined the inflation and unemployment and interest rate in Sri unemployment nexus in Nigeria by testing if Lankan economy. the original Phillips curve proposition holds iv. To examine the short run relationship for Nigeria. The study adapted a distributed among the variables, inflation, lag model with data covering the period 1970- unemployment and interest rate in Sri 2011. Blanchflower, et al. (2014) studied on Lankan economy. the happiness tradeoff between unemployment and inflation. This paper Methodology makes use of a large European dataset, This study purely depends on the secondary covering the period 1975 to 2013, to estimate data. In an attempt to explore empirically the happiness equations in which an individual relationship between inflation, subjective measure of life satisfaction is unemployment and interest rate in Sri Lanka, regressed against unemployment and a multiple regression model such as ARDL inflation rate. Touny (2013) set the main was employed. According to the literature, objective of his study was to investigate the this model includes inflation (INF), long run trade-off between unemployment unemployment (UR), interest rate (IR), and inflation in Egypt through the period Money Supply (M2) and government (1974-2011).The study revealed a positive expenditure (GE) as variables; where the relationship between changes in inflation rate inflation rate is dependent variable and others and unemployment gap in the long run. Van are explanatory repressors. Logarithmically Bon Nguyen (2015) studied about effects of transforming variables in a regression model fiscal and money supply (M2) on inflation: is a very common way to handle situations Evidence from selected economies of Asia where a non-linear relationship exists and found that money supply (M2) has between the independent and dependent significantly positive relationship with variables. Using the logarithm of one or more inflation Mohseni and Jouzaryan (2016) variables instead of the un-logged form studied on the subject of examining the makes the effective relationship, while effects of inflation and unemployment on preserving the linear model. Necessary economic growth in Iran for the period of secondary data collected from published and

3 unpublished data sources, such as books, contribution of the performance of journals, proceedings, newspaper articles, unemployment and interest rate on inflation statistics, database and handbook, various in Sri Lanka. The basic model constructed reports of central bank, world bank and IMF, was university research pool, web browsers and other bibliographies related to the content. 퐼푁퐹푡 = 훽표 + 훽1푈퐸푡 + 훽2퐼푅푡 + 훽3푀2 + 훽4퐺퐸푡 + 휀푡 … (1)

The log- log model was constructed by considering The data and information collected from data type and trend. Using log transformation, the secondary sources analysed using various model is specified as follows: analysis methods. The statistical tools like average, percentage, and parametric and non- 퐿표푔 퐼푁퐹푡 = 훽표 + 훽1푙표푔푈퐸푡 + 훽2푙표푔퐼푅푡 + 훽3푙표푔푀2 + parametric tests used where necessary. First 훽4푙표푔퐺퐸푡 + 휀 … … … … … … … … … … … … … … … … … . … (2) of all, non-parametric approach or graphical Here, the model re-organized and named as follows for explanation such as Kernel Fit with the analysis purpose Confidence ellipse curve was derived among the variables and shown the relationship. The ln퐼푁퐹푡 = 훽표 + 훽1ln푈퐸푡 + 훽2ln퐼푅푡 + 훽3ln푀2 + 훽4ln퐺퐸푡 non parametric relationship was validated + 휀 … … … … … … … … … … … … … … . (3) using parametric approaches such as co- Where INF, UE, IR, GE and M2 are denoted as integration technique. The co-integration inflation rate, unemployment rate, interest rate, technique was employed to establish Government Expenditure and Money supply (M2) relationships between variables in short run respectively,  is the error term and are parameters. The log denoted by and long run as well. The direct effects of 0 ,1,2 ,3and4 these inflation, interest rate and ln or LN in the analysis. The best model for analysis unemployment could be identified easily this time series data is selected by using the result rather than the indirect effects or relationship. obtained from unit root test. The figure 1 below clearly explains the situation of data stationary and To identify the time series properties of the appropriate models to be selected. variables, Dickey –Fuller (DF) test with break unit root technique was employed. According Statistical model selection based on stationary of the data to the DF test of stationary, the ARDL model was selected and analysed to determine the Stationary Stationary at Stationary at Stationary at st st nd relationship between the variables in Sri at level level & 1 1st Difference 1 & 2 Lanka. Co-integration technique was used to Difference Difference examine the short and the long run Auto relationship between the variables. Short run Regression ARDL Co-integration Regressive and long run relationship as well as the long models run equilibrium of the model were estimated by using the ARDL co-integration model. Unrestricted VAR Error Correction The trend of variables such as inflation, interest rate and unemployment explains by using the tables and graphs. The software ECM - One VECM – More than one endogenous variable endogenous Variable EViews version 9. and Microsoft Excel used for this purpose. Data is displayed in time Figure 1. Model selection series graph at various time-points. This is Source: Muhammad Saheed Aas khan Meo, 2016 another type of graph used for this kind of specific data that come in pairs. The vertical Result and Discussions axis is for data values while the horizontal The trend of variables such as inflation, axis shows time. This kind of graph used for unemployment, interest rate, money supply showing trends passing through a time (M2) and government expenditure for the period. The following analytical function sample period, figured out in following given below is used to test the data on the graphs.

4

30

20

10 (%) 0 Inflationrate 1960 1970 1980 1990 2000 2010 2020 -10 Year Inflation Figure 2. The trend of inflation in Sri Lanka, 1963 - 2015

25.00

20.00

15.00

(%) 10.00

5.00

Unemployment Rate 0.00

1960 1970 1980 1990 2000 2010 2020

Year Unemployment rate

Figure 3. Trend of unemployment rate in Sri Lanka, 1963 - 2015

30.0

20.0

10.0

Interest Rate (%) 0.0 1960 1970 1980 1990 2000 2010 2020 Year Interest rate Figure 4. Trend of interest rate in Sri Lanka, 1963 - 2015

5

4

3

2

1

Money Money Supply (M2) 0 1960 1970 1980 1990 2000 2010 2020 Year Money Supply (M2)

Figure 5. Trend of money supply (M2) in Sri Lanka, 1963 - 2015

50.00

40.00 30.00

20.00

10.00

0.00 Govt. Expenditure 1960 1970 1980 1990 2000 2010 2020 Year Govt Expenditure

Figure 6. Trend of government expenditure in Sri Lanka, 1963-2015

5

LGE -3.551842 -8.444567a -12.98718a -12.61293a 28 a a LM2 -1.485458 -4.720611 -9.740912 -9.870845 24 Source: Estimated, 2017 (a - 1% significance, b -, 5% 20 - & c - 10%) 16

n 12 d

o i b The Dickey-Fuller (DF) unit root tests with

t

a

l

f

n 8 I c break unit root technique is performed on e 4 a both levels and the first differences with

0 intercept or intercept and trend for all the

-4 variables. The result of the unit root test

-8 (Dickey Fuller) for the variables are 0 2 4 6 8 10 12 14 16 18 20 22 24 26 presented in the table 1 above. The Dickey Unemployment Fuller Test results confirm that the time series data of some variables in the model are non- Kernel Fit 0.95 Ellipse stationary in their levels. However, these Figure 7. Visual inspection of the relationship variables are stationary in their first between inflation & unemployment difference. Which means, some variables are

integrated I (0) series and some variables are The Kernel Fit with Confidence ellipse graph integrated I (1) series. But, all the variables 7 above shows the relationship between the are stationary at I (1) which means all the inflation and unemployment rate in Sri variables are integrated at 1st difference with Lanka. It fluctuates over the samle period of intercept. According to the model selection time and shows an up and down movement. criteria the ARDL (Auto Regressive Between the point a-b and c-d show a positive Distributed Lag) model is most suitable for trend and b-c and d-e show a negative trend this kind of time series data. Hence, the rest between the variable. The graph 7 is not of the analysis carried out by using the ARDL showing a relationship among the variables model. clearly, but it shows a relatively negative trend. Bound test: Bound test is performed in order to find the long run relationship of the To meet one of the objectives of the study and variables. If the F statistic is greater than the find impact of unemployment and interest critical Value of particular confidence level, rate on inflation, the DF break unit root test it means that the null hypothesis stated ‘No was employed. Unit root test is essential to long run relationship exit’ is rejected which find out proper estimation model. The means there is a long run relationship nonparametric or graphical approach shows between the variable is exist. According to the that the variables are fluctuated and break. statistic mentioned in the table 2 below, the F Hence, the break points unit root test statistic is 8.395 and upper bound of critical employed in order to find the real impact of value for 1% confidence level is 4.37. The variables. result clearly shows that, even at 1% of critical level of upper bound is exceeded. Table 1. Result of break unit root test Hence, the model is suitable for the DF at level DF at 1st Difference cointegration test. According to the residual diagnostic tests mentioned above the model is Variable Intercept Intercept Intercept Intercept & trend & trend fitted and specified correctly. Consequently, LINF -2.039679 -2.633900 -7.464779a -7.637685a the short run and long run relationship of the

c a a independent variables with inflation could be LUE -4.206792 -4.396421 -8.575309 -8.142210 established. LIR -4.346063c -3.951242 -9.711126a -9.719216a

6

Table 2. Test statistic of bound test 20 Test Statistic Probability 15 Bound Test 8.395 0.971 10

Critical value 5 0 Bound 1 Bound 0 a. Significance at 3.29 4.37 -5 1% b. Significance at 2.56 3.49 -10 5% -15 c. Significance at 2.2 3.09 -20 10% 86 88 90 92 94 96 98 00 02 04 06 08 10 12 14 Legend: a: Significant at 1%, b: Significant at 5%, CUSUM 5% Significance c: Significant at 10% Figure 9. CUSUM test 1.4 Source: Estimated, 2017 Tests of model stability :The graph and the 1.2

ARDL regression result show there is an 1.0 impact of unemployment on inflation in Sri 0.8

Lanka. Though the statistical result confirms 0.6 the phillip’s relationship among the variables, 0.4 it should be re-checked whether the selected 0.2 model is fulfilling the criteria of model 0.0 stability tests in order to further discussion. -0.2 -0.4 Schwarz Criteria (top 20 models) 86 88 90 92 94 96 98 00 02 04 06 08 10 12 14

2.04 CUSUM of Squares 5% Significance 2.02 Figure 10. CUSUM square test

2.00 Source: The test parameter finds instability if the 1.98 cumulative sum goes Estimated, outside 2017 the area 1.96 between the two critical lines. The graphs 9 1.94 & 10 above clearly show that the model fitted

1.92 within the confident level line. Hence, the

1.90 model used for the analysis is best based on

) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )

4 4 4 4 4 4 4 4 4 0 4 4 4 1 4 4 4 4 4 2

, , , , , , , , , , , , , , , , , , , , the CUSUM test graph.

2 2 3 3 2 2 1 3 1 2 2 3 2 2 2 2 4 3 3 2

, , , , , , , , , , , , , , , , , , , ,

1 2 1 2 3 1 1 0 2 1 2 3 1 1 0 2 1 1 1 1

, , , , , , , , , , , , , , , , , , , ,

4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

, , , , , , , , , , , , , , , , , , , ,

1 1 1 1 1 2 1 1 1 4 2 1 3 4 1 3 1 3 2 4

( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( (

L L L L L L L L L L L L L L L L L L L L

D D D D D D D D D D D D D D D D D D D D Ramsey RESET test: There are many

R R R R R R R R R R R R R R R R R R R R A A A A A A A A A A A A A A A A A A A A statistical techniques to test the model Figure 8. Model selection of Schwarz specification. The Ramsey Regression

criterion Equation Specification Error Test (RESET) is According to the Schwarz criteria, 2500 tested the specification of independent models evaluated and selected the ARDL (1, variables. This omitted variable test, namely 4, 1, 2,4) is the best model for the data set. Ramsey RESET test was employed for the The graph 7 above shows the top twenty best understanding of whether there are any more models and explained that the ARDL (1, 4, 1, variables to be added to the betterment of 2,4) is carrying a minimum Schwarz criterion analysis or the included variables to be (SIC) which the best one. specified as square or cube format. The

hypothesis of the Ramsey RESET test is as CUSUM test : The CUSUM test (Brown, follows, 퐻 The model is correctly specified, Durbin, and Evans, 1975) is based on the 0: 퐻 The model is not correctly specified. cumulative sum of the recursive residuals. 1:

This option plots the cumulative sum together with the 5% critical lines.

7

Table 3. Result of Ramsey RESET test bell-shaped and the Jarque-Bera statistic Omitted Variables: Powers of fitted values from 2 to 5 should not be significant for a discussion of Value df Probability the Jarque-Bera test The graph 10 shows as F-statistic 1.407571 (4, 27) 0.2582 bell-shaped and the Jarque-Bera statistic is F-test summary: not significant at 5% confidence level (JB is Mean 0.059 and p is 0.971). Hence, the used model Sum of Sq. df Squares is specified correctly as per the Jarque-Bera Test SSR 0.842210 4 0.210553 test statistic. Restricted SSR 4.881026 31 0.157452 Unrestricted SSR 4.038815 27 0.149586 According to the results shown in graph 10 The F statistic of the Ramsey RESET test is above and table 4 above, the error term of the 1.407 with a probability of 0.2582 which model in this study normally distributed and means that the null hypothesis of RESET test it is statistically not significance too. This specified above cannot be rejected at 5% study concludes that the specified model has confidence level. Hence, the model no autocorrelation, no heteroskedasticity, developed in this study is correctly specified well specified functional form and the and there is no more square or cube form of regressions is stable. Therefore, it can be independent variables to be added. concluded that the model applied in this study is robust and the specification of the model is Jarque-Bera (Normality test): This below adequate. Hence, rest of this study analysed graph 10 and table 4 below clearly display a using specified ARDL model with histogram and descriptive statistics of the cointegration technique in short run and long residuals, including the Jarque-Bera statistic run relationship of variables in Sri Lanka. for testing normality. If the residuals are normally distributed, the histogram should be

Figure 11. Result of Jarque-Bera normality test

Table 4. The results of diagnostic tests based on the model Test Statistic Probability Decision

Jarque-Bera (Normality test) 0.0586 0.971 Error Normally distributed Heteroskedasticity Test No heterokedasticity 1. Breusch-Pagan-Godfrey 1.184 0.332 2. ARCH Test 0.588 0.447 No heterokedasticity Breusch-Godfrey Serial Correlation 1.096946 0.3781 No serial Correlation LM Test: Legend: a: Significance at 1%, b: Significant at 5%, c: Significant at 10%

8

According to the results shown in graph 10 inflation at 5% confidence level. Interest rate above and table 4 above, the error term of the and money supply (M2) are negatively model in this study normally distributed and correlated with inflation. A unit change in the it is statistically not significance too. This interest rate and money supply (M2) will study concludes that the specified model has negatively change the inflation by 1.844 and no autocorrelation, no heteroskedasticity, 4.138 times respectively with statistical well specified functional form and the significant at 5% confidence level. regressions is stable. Therefore, it can be Government expenditure is positively concluded that the model applied in this study correlated with inflation and a unit increase in is robust and the specification of the model is the government expenditure will increase the adequate. Hence, rest of this study analysed inflation by 0.846 times. But the government using specified ARDL model with expenditure is not statistically significant at cointegration technique in short run and long 5% confidence level. According to the short run relationship of variables in Sri Lanka. run Co-integration regression output presented in the above table 5.8 shows that, Short run relationship: The short run out of four independent variables, three relationship of the variables were analysed variables and its lag values are statistically using co-integration technique. The following significant even at 5% confidence level. The co-integration regression result reveals that critical value and p value of the model is - the relationship between inflation and 0.709 and 0.000 respectively. The model is respective explanatory variables. Some overall significant even at 1% confidence variables show the negative relationship and level. Hence, as per the short run analysis the rest show a positive relationship with described above, this concludes that the inflation in the short run. unemployment and interest rate negatively correlated with inflation for the sample period Short run co-integrating equation is as in Sri Lanka. follows: Long run relationship: The long run co- 퐼푁퐹 = −2.124 − 0.147푈퐸 − 1.844IR − 4.138M_2 + 0.846퐺퐸. integration test estimated and the following -0.709 (0.000) calculated long run co-integrating equation is derived to meet the purpose of this study. Table 5. ARDL Co-integration regression results for the model Coefficient & INF = −2.114 − 0.384 UE + 0.420 IR + 0.045 MS(M_2 ) + 1.248 GE Variable (0.0262) (0.0142) (0.8462) (0.3091) Prob. b Unemployment -0.147 Further, the estimated coefficient of (0.037) Interest rate -1.844a unemployment indicates that, when other (0.009) variables are constant a unit increase in a Money Supply (M2) -4.138 unemployment rate will reduce 0.3845 times (0.003) of inflation. The long run relationship Govt. Expenditure 0.846 between inflation and unemployment is (0.223) negative and statistically significant at 5% Legend: a: Significant at 1%, b: Significant at 5%, c: Significant at 10% confidence level. The result clearly indicates with statistical witness that the Phillip’s The table 5 above shows estimated result of relationship of variables exist in short run and short run co-integration regression, which long run in Sri Lanka. The result confirmed reveals that when other variables are constant the finding of Balamurali and Sivarajasingam a unit increase in the unemployment rate will (2013) in Sri Lanka. The estimated decrease the inflation by 0.147 times. coefficient of interest rate indicates that, a Unemployment is negatively correlated with unit increase will increase the inflation at a

9 rate of 0.42 times. The long run relationship These two variables have an important between inflation and interest rate is positive role in the socio- economic development. and statistically significant at 5% confidence A certain level of inflation also needed to level. Also, the money supply (M2) and encourage the investors to supply more government expenditure are positively products to the market. But, it should be correlated with inflation and a unit increase of under controlled observation because money supply (M2) and Government these variables show a negative Expenditure will increase the inflation at a relationship in Sri Lanka. rate of 0.545 & 1.248 times respectively. . The relationship between variables Here, money supply and government interest rate and inflation is negative in expenditure are statistically not significant at short run and positive in the long run. 5% confidence level. Policy makers have to utilize this relationship in order to control the Conclusion inflationary situation. Once the inflation, The impact of unemployment and interest increasing continuously, policy decisions rate on inflation in Sri Lanka was analysed could be taken to increase a certain level and discussed. The results were found of interest rate in order to reduce the presented in this chapter. The parametric and inflation. But, this is not a long term non-parametric approaches were employed in policy instrument for the inflationary order to find the trend and the relationship of situation in Sri Lanka because the positive the variables. The kernel fit and confident relationship among the variable found in ellipse curve showed a negative relationship the long run. among the variables inflation and . The fiscal consolidation and reform in unemployment. The unit root test of DF was public enterprises by reducing employed to identify the stationary property considerable wasteful public expenditure of the variables and decided to run the ARDL are to be needed in order to maintain a model since the unit root test fulfilled the sustainable development. According to properties of ARDL. The ARDL (1,4,1,2,4) the finding, the government expenditure model comprised with inflation as dependent has a positive relationship with inflation and unemployment rate, interest rate, money in Sri Lanka. Therefore, the national supply (M2) and government expenditure are policies on economy and development as independent variables and evaluated by have to be implemented with optimum using graphs and statistics to conclude the benefits. scenario. The graphical approach shows relatively negative trend between inflation . The export sector initiatives of Sri Lanka and unemployment but it is not clear. Then, have to be strengthened and exporter to be the statistical approach namly ARDL co- motivated to have a better balance of integration technique used and evaluated the payment. Attractive tax reliefs to be log-log ARDL model with the lag values of introduced for new start-up businesses the variable. Finally, the long run relationship and rural based export initiatives to be evaluated and found that the impact of encouraged. The government of Sri unemployment on inflation is negative in the Lanka has to promote policies that inspire short run & long run as well. Also, the impact investor with confidence by ensuring law of interest rate on inflation is negative in and order. Then only, the job short run and positive in the long run. opportunities in the private sector rise in order to create a job market for Recommendations unemployed. . The optimum trade-off between inflation . The government investment has to be and unemployment has to be maintained. pumped more and more on the public and

10

private enterprises to create the job . The free flow of information between opportunities. The government initiatives employers and employees should be on job creating activities should be enhanced. Training and educational strengthened. Most of the unemployed are programmes should be increased and expecting job in the public sector (Paulina geared towards innovations and Mary & et al, 2014) and the government productivity in order to reduce the rate of also formulate the policies, not to create unemployment in the economy. the job opportunities in the private sector and providing jobs in the public sector by expecting votes in election. This trend References should be evaluated and the innovative Adeyi emmanuel ola. (2012). Unemployment policies on job creation to be and Inflation in Nigeria: An empirical implemented. Also, the expectations of investigation, Sri Lanka economic job seekers in Sri Lanka have to be research conference, University of corrected in positive way through the Peradeniya. education policies of Sri Lanka. Alfred A. Haug & Ian P. King (2012). . The Sri Lankan economy structurally Empirical Evidence on Inflation and changed from agriculture to service since Unemployment in the Long Run, [online]. the independent.But, the industrial sector Available http://papers.ssrn.com/ is still in their early life cycle and it should sol3/papers.cfm?abstract_id=1926475 be developed in order to create more job [Accessed: 14/11/2015] opportunities in the private sector. Also, Anthony Orji., Onyinye, I. & Joan, C. O. the industrial development has to be (2015) , The inflation and unemployment extended out of in order to nexus in Nigeria, Asian Economic and equalize the development phase Financial Review, Asian Economic and throughout the country. Environmental Social Society (AESS) 5(5): pp.766-778. concern is to be top of the mind when Balamurali, N. and Sivarajasingam, S. (2013) industrial development taking place. Examining the Trade-off between Inflation and Unemployment Rate in the . In Sri Lanka, there are no more five or ten Long Run in Sri Lanka: Parametric and years long term economic plans Non-Parametric Econometric implemented continuously. The Investigations. In: 2nd International government of Sri Lanka implemented its Economics Research Conference. 13th own economic policies time to time. and 14th December 2013. Department of Suppose the government changes, the Economics and Statistics, University of implemented economic policy will also Peradeniya, Sri Lanka. change according to the decision of ruling Blanchflower D. G, Bell, D.N.F,Montagnoli, political party. But, this is the time to A and Moro, M. (2014). The happiness rethink about the long term economic tradeoff between unemployment and policies and create an administrative inflation, Journal of Money, Credit and division called an engine of economic Banking, Stirling Management School, growth which should carry out the task University of Stirling. continuously to implement the economic Central Bank of Sri Lanka. (1998) Economic policies throughout the period with the Progress of Independent Sri Lanka. adaptation of innovative ideas. Even Colombo: Central Bank of Sri Lanka. though, the government changes the Chaido, D. and Melina D (2012).Inflation, permanent long term economic policies Unemployment and the NAIRU in Greece, should not be changed. As like in India, Elsevier Ltd publication, International China and Malaysia, continous five year Conference On Applied Economics, plans have to be implemented.

11

Procedia Economics and Finance 1 Mohseni, M. and Jouzaryan, F. (2016) (2012) 118 – 127 Examining the effects of inflation and Department of Census and Statistics (2014). unemployment on economic growth in Labour Force Survey annual report 2014 Iran, 1st International Conference on [online]. Available Applied Economics and Business, http://www.statistics.gov.lk/samplesurve ICAEB 2015, Procedia Economics and y/LFS_Annual%20Report _2014.pdf Finance 36 ( 2016 ) 381 – 389. [Accessed: 20/12/2015].Ewing, B.T and Niranjala, S.A.U. (2014) The Relationship Seyfried, W.L. (2004). A time-varying Between Inflation And Unemployment: volatility approach to modeling the A Study Based on Sri Lankan Economy. phillips curve: A cross-country analysis, Proceedings of the Wayamba University In: Journal of Economic and Finance, International Conference, Sri Lanka, 29 - June 2004, Vol. 28-2 (2003) pp 186-197 30 August 2014. p 134. Gulistan, E., Hasan, G, D. & Gungor, K. OpenStax College. (2014), Principles of (2015). The Analysis of The Relationship Economics. OpenStax College. 19 March Between Unemployment And Inflation In 2014. [online]. Available Turkey By VAR Model, Journal of new http://cnx.org/content/col11613/latest/ results in Science, Graduate school of [Accessed:14/12/ 2015] Natural & Applied science, university of Phillips, A.W. (1958). The relationship Gaziosmanpasa Vol :8, Pages: 22-29 between unemployment rate and the rate Handapangoda,W,S.(2005).The relationship of change of money wage rates in the between price (Inflation) and United Kingdom 1961-1957, Economica, Uuemployment;Philips Curve- A Based the London school of economics, pp 283- on the Sri Lankan Economy,Proceding of 299. International Conference on Business Touny. M. A. (2013) Investigate the long run Management. trade-off between unemployment and http://journals.sjp.ac.lk/index.php/icbm/a inflation in Egypt, Canadian Center of rticle/view/1081 Science and Education , International Hussein Ali Al-Zeaud (2014) The Trade-Off Journal of Economics and Finance; Vol. between Unemployment and Inflation 5, No. 7, pp 115-125. Evidence from Causality Test for Jordan, Van Bon Nguyen. (2015) Effects of fiscal and International Journal of Humanities and money M2 supply on inflation: Evidence Social Science Vol. 4 No. 4, Special Issue from selected economies of Asia, Journal – February 2014, pp 103 – 111. of Economics, Finance and ILO, (2016), Department of statistics [online] Administrative Science, Faculty of Public 21st May 2015. Available from: Finance, University of Economics Ho Chi http://www.ilo.org/public/english/bureau Minh City, Vietnam, pp.49–53. /stat/res/index.htm World Bank (2015). Gross National Income Lakshman, W.D. (2004) Youth Per Capita 2014, Atlas Method and PPP Unemployment: An Exploratory Study. [online]. Available Essays in Economic Policy in Sri Lanka. http://databank.worldbank.org/data/down Colombo. Vijitha Yapa Publication. load /GNIPC.pdf [Accessed: 14/11/2015] Lalani, J.M.G. (2014). The Study of Testing Validity of Phillips Curve for Sri Lanka. In: 2nd International research symposium. 29-30th October, 2014. Mihinthale. Faculty of Social Sciences and Humanities Rajarata University of Sri Lanka.

12