tiểu luận kinh tế lượng factors affecting unemployment rate in japan

FOREIGN TRADE UNIVERSITY INTERNATIONAL ECONOMICS FACULTY ****************** ECONOMETRICS FINAL EXAM TOPIC Factors Affecting Unemployment Rate in Japan Class : 57 – JIB Lecturer : Assoc Pro Dr Tu Thuy Anh Group member : Phan Thi Thu Trang - 1815520236 Le Thuy Linh - 1815520188 Tran Thuy Huyen - 1815520181 Ha Noi, 10/2019 Table of Contents I Introduction………………………………………………………………………… Object……………………………………………………………… ……………… Scale of the research……………………………………………………………… II Theorical background………………………………………………………….……4 III Literature review……………………………………………………………………8 IV Data description………………………………………………………… …………9 Variable table…………………………………………………………… Data description……………………………………………………………………….9 Correlation matrix………………………………………………………….……… 10 V Econometrics model………………………………………………………… …….10 Population Regression Function…………………………………………………….10 Sample of Regression Function………………………………………………… ….10 Result………………………………………………………………………………….11 Meaning of Coefficient……………………………………………………………….11 Testing hypothesis relating to a regression coefficient………………………………12 VI Robustness check………………………………………………………………… 14 Multi-collinearity……………………………………….……………………… … 14 Heteroskedasticity……………………………………………………………… … 15 Normality………………………………………………………………………… 16 Auto correlation…………………………………………………………………… 17 VII Cure…………………………………………………………………………… …18 VIII Recommendation……………………………………………………………… 20 IX Conclusion……………………………………………………………………… 21 X References………………………………………………………………………… 22 XI Appendix………………………… …………………………………………… …23 I Introduction Econometrics is defined as the social sciences in which the tools of economic theory, mathematics and statistical speculation are applied to the analysis of economic problems Econometrics analyzes real - world data using statistical methods to test or develop economic theory as well as to more understand those issues Through the results, econometricians can make predictions of economic phenomenon Unemployment is a social problem that both developed and developing country are considered As the general, more unemployed workers mean less total economic production will take place than might have otherwise And unlike idle capital, unemployed workers will still need to maintain at least subsistence consumption during their period of unemployment This means the economy with high unemployment has lower output without a proportional decline in the need for basic consumption High, persistent unemployment can signal serious distress in an economy and even lead to social and political upheaval On the other hand, a low unemployment rate means that the economy is more likely to be producing near its full capacity, maximizing output, and driving wage growth and rising living standards over time Japan - a developed country is known as the third largest economy And Vietnam - a developing country has high the number of labor export to developed country like Japan According to the statistic data from the World Bank (2018), the unemployment rate in Vietnam and Japan is 1,891% and 2,335% respectively As the theory above, the unemployment rate in Japan is supposed to be lower than the unemployment rate in Vietnam As economics students, we recognize the need to study and research about Econometrics in logical and problem analysis as well as interested in Japan’s labor market So, to better understand how to apply the Econometrics into reality, our group which includes three members: Phan Thi Thu Trang, Le Thuy Linh and Tran Thuy Huyen would like to develop Econometrics report (known as final exam) under the guidances of Assoc.Prof Dr.Tu Thuy Anh and PhD Chu Thi Mai Phuong In this report, we used the econometric analysis tool GRETL to analyze the topic "Factors Affecting Unemployment Rate in Japan” Object The object of the topic is analyze the influence of Gross Domestic Product Growth Rate, Population Growth Rate, Inflation Rate and Foreign Direct Investment to unemployment problem in Japan (measured by Unemployment Rate variable) The report tends to concentrate on directly objectives below: - Revising the knowledge that we have studied through the course and also have the chance to test of the influence of GDP Growth Rate, Inflation Rate, FDI & Population Rate to Unemployment Rate in the world, specially in Japan - Using Ordinary Least Squares Regression to analyze the affecting of variables to Unemployment Rate Testing and fixing errors of model estimated Then, giving suggestion and solution for the problem from real data Scale of the research Researching about the affecting of factors: GDP Growth Rate, Population Growth Rate, Inflation Rate and FDI to Unemployment Rate in Japan from 1980 to 2018 Finally, we want to give a sincere thank for our instructors - Assoc Prof Dr.Tu Thuy Anh and PhD Chu Thi Mai Phuong for helping us to implement this report Through this assignment, we had the chance to review, consolidate and use the knowledge gained from the Introduction to Econometrics Course to analyze a real issue Despite all the efforts, we certainly can not avoid the errors, we look forward to your comments so that our team can improve this report II Theorical background Unemployment rate 1.1 Definition The unemployment rate is defined as the percentage of unemployed workers in the total labor force Workers are considered unemployed if they currently not work, despite the fact that they are able and willing to so 1.2 Calculating Unemployment Rate x 100% 1.3 Unemployment theory a Classical Unemployment The term classical unemployment refers to the effect real wages and the market clearing wages has on the availability of jobs Classical unemployment increases when real wages rise relative to the market clearing wage Also known as real-wage unemployment, increases and decreases in classical unemployment is a function of the law of supply and demand Classical unemployment occurs when the wages a worker is willing to accept (real wages) is in excess of those an employer is willing to pay (market clearing wage) By definition, the market clearing wage is the equilibrium wage That is to say, it is the wage at which the supply of labor is equal to the demand for the labor This relationship is illustrated below: At the market clearing wage (MCW), the demand for labor is exactly equal to the supply of labor and classical unemployment would be zero When real wages (RW) are in excess of the market clearing wage (MCW), classical unemployment is greater than zero The Philips Curve: The relation between inflation rate and unemployment rate The idea for the Philips curve was proposed in 1958 by economist A.W.Philips and he found that there was a stable, inverse relationship between wages and unemployment Then in 1960, economists Paul Samuelson and Robert Solow expanded this work to reflect the relationship between inflation and unemployment The Philips curve argues that unemployment and inflation are inversely related: levels of unemployment decrease, inflation increase Graphically, the short-run Philips curve traces an L-shape when the unemployment rate is on the x-axis and the inflation rate is on the y-axis The theory of the Philips curve seemed stable and predictable Data from the 1960’s modeled the trade - off between unemployment and inflation fairly well The Philips curve offered potential economic policy outcomes: fiscal and monetary policy could be used to achieve full unemployment at the cost of higher price levels or to lower inflation at the cost of lowered employment However, when governments attempted to use the Philips curve to control unemployment and inflation, the relationship fell apart Data from the 1970’s and onward did not follow the trend of the classic Philips curve For many years, both inflation rate and unemployment rate were higher than the Philips curve would have predicted Keynesian unemployment theory Keynes developed his theories in response to the Great Depression, and was highly critical of classical economic arguments that natural economic forces and incentives would be sufficient to help the economy recover In Keynesian Unemployment, it is the situation where low wage-rates should result in higher employment levels, but don’t because the economy is in recession and the employees are facing low demands for their goods and services According to Keynesian theory, changes in aggregate demand, whether anticipated or unanticipated , have their greatest short-run effect on real output and employment, not on prices Keynes economics has pointed out that prices and wages are flexible, full-employment is hard to achieve and not sure to be good It means that the economy needs to balance the wage employees want to receive and the wage employers willing to pay Keynesian theory of unemployment is a demand-deficient theory This means that Keynes visualized unemployment from the demand size of model His theory, also known as demand-oriented approach, opposed to classical supply side model According to Keynes, the volume of employment in a country depends on the level of effective demand of people for goods and services Overview about factors affecting to unemployment rate 4.1 Gross domestic product growth rate Gross Domestic Product (GDP) is a moneytary measure of the market value of all the final goods and services produced in a specific time period (according to wikipedia) Economic growth is measured by gross domestic growth rate Theoretically, when a country has higher growth means unemployment is soluted This thing shows the inversely relationship between economic growth and unemployment rate In economics theory, Okun’s law (Aurthur Melvin Okun) (1962) proposed the relationship the relationship between unemployment and losses in a country's production which was summed up through his observation Okun’s law pointed out the approximately estimation that “2% increase in output corresponds to a 1% decline in the rate of cyclical unemployment; a 0.5% increase in labor force participation; a 0.5% increase in hours worked per employee; and a 1% increase in output per hours worked” 4.2 Inflation rate A.W.Philips is one of the first economist found the method to prove the inversely relationship between inflation and unemployment This correlation is showed through Philips curve (1958) William Philips supposed that unemployment rate and inflation has inversely relationship It means that if a economy want to have low unemployment rate, its economy must 4.3 Foreign direct investment growth rate Foreign Direct Investment (FDI) is an investment in the form of a controlling ownership in a business in one country into business by an entity based in another country (according to wikipedia) FDI is one of the important resources to compensate for foreign investment, promotes economic developmen to countries invested, specially to developing and under-developed countries It is an basically channel for the transfer of technology between countries, promotes international trade In reality, FDI impacts directly both positive and negative on unemployment rate in different aspects 4.4 Population growth rate Population Growth Rate (PGR) is the changing of population for a period of time According to Malthusian theory of population (1798), the increase in population under controlled will cause the higher of unemployment Other words, the vast population causes higher unemployment rate in any economy III Literature review Saga Katria, The study of the relationship between inflation and unemployment with eight regional member countries of SAARC and six expected future member countries from the perspectives of Philips curve for the period 1980 - 2010 The results shows that there is trade-off between inflation and unemployment Relationship between unemployment and the inflation rate in India (Philips curve) The study is based on secondary data collected from planning commission of India; inflation.edu.in and the ordinary least square & simple linear regression has been used for analysis of data The results suggests that there is a positive relationship between inflation and unemployment Therefore, the Philips curve did not come true in the context of India’s economy O’Nwachukwu (2017), Determinants of the Rate of Unemployment in Nigeria The study examines the determinants of unemployment rate in Nigeria from 1980 to 2016 And the researcher employed the Ordinary Least Squares (OLS) method to estimate the model after using the Augmented Dickey-Fuller to test for unit root And the results is first lag of unemployment in Nigeria and GDP were not statistically significant in explaining unemployment in Nigeria Dr Aurangzeb, Khola Asif (January 2013), Factors affecting Unemployment: A cross country analysis This study investigates macroeconomic determinants of the unemployment for India , China and Pakistan for the period 1980 to 2009, using co integration, granger causality and regression analysis , The variables selected for the study are unemployment, inflation, gross domestic product, exchange rate and the increasing rate of population The results of regression analysis showed significant impact of all three variables for all three countries It is recommended that distribution of income needs to be improved for Pakistan in order to have positive impact of growth on the employment rate IV Data description Variable table Variables Abbreviation Meaning Unit Unemployme Y nt rate The proportion of the labor force that is not % currently employed but could be GDP growth X1 rate The change in a Japan’s gross domestic % The change in Japan’s population % Population growth rate Inflation rate X2 X3 The percentage at which a currency is devalued % during a period FDI net X4 inflows (% of GDP) The value of inward direct investment made by % non-resident investors in the reporting economy Data description Mean Median S.D Min Max Unemployment rate (Y) 3.472 3.400 1.047 2.000 5.400 GDP growth rate (X1) 1.955 1.663 2.244 -5.416 6.785 Population growth rate (X2) 0.2257 0.2380 0.2839 -0.2030 0.7840 Inflation rate (X3) 1.001 0.5960 1.752 -1.353 7.779 FDI net inflows (X4) 0.1447 0.07100 0.1925 -0.05300 0.8310 Description: Urate Mean: The average unemployment rate of 39 observations is 3.472% Urate Median: Fitted value of dependent variable unemployment rate is 3.4% Urate Minimum: The minimum unemployment rate among 39 observations is 2% Urate Maximum: The maximum unemployment rate is 5.4% Std Dev (Standard Deviation): Is a measure of how spread the numbers are, equals to the square root of sample variance The Std Dev of unemployment rate here is 1.047 Correlation matrix Urate GDP growth rate P growth rate I rate FDI 1.000 -0.5140 -0.5176 -0.7197 0.2087 U rate 1.0000 0.5406 0.3641 -0.4003 GDP growth rate 1.0000 0.5699 -0.5413 P growth 1.0000 -0.2387 I rate 1.0000 FDI Look at the table of correlation, we see that correlation of almost the independent variables with each others are low V Econometrics model Population Regression Function (PRF) Y = �0 + �1.X1 + �2.X2 + �3.X3 + �4.X4 + ui Sample of Regression Function (SRF) Y = �0 + �1.X1 + �2.X2 + �3.X3 + �4.X4 + ei (ei is error) 10 Result R2=0.596478: It means that the regressors explain 59.64% of the variance of unemployment rate SER = 0.702813: It estimates standard deviation of error ui A relative low spread of scatter plot means that prediction of unemployment rate base on these variables might be reliable We have the temporary econometrics model for unemployment rate in Japan: Y = 4.22127 + (-0.138534).X1 + (-0.239892).X2 + (-0.355520).X3 + (-0.475336).X4 + e Meaning of Coefficient �0: If all these other factors equal to zero, the unemployment rate equals to 4.22127% �1: If the GDP growth rate increases by 1%, the unemployment rate will decrease by 0.138534% �2: If the price growth rate increases by 1%, the unemployment rate will decrease by 0.239892% �3: If the inflation rate increases by 1%, the unemployment rate will decrease by 0.355520% �4: If the FDI increases by unit(?), the unemployment rate will decrease by 0.475336% 11 Testing hypothesis relating to a regression coefficient 5.1 Intercept β0 Null hypothesis: H0: β0=0 Alternative hypothesis: H1: β0≠ The student t distribution 5% (2 sides) when n=39 ts=1,96 We have: tob < ts p-value=4.31e-019 At 5% level of significance, we have enough evidence not to reject H0 => linear relationship is recorded between X2 and Y 12 5.4 Coefficient β3 Null hypothesis: H0: β3=0 Alternative hypothesis: β3 ≠ The student t distribution 5% (2 sides) when n=39 ts=1,96 p-value=8.88e-05 At 5% level of significance, we have enough evidence to reject H => linear relationship is recorded between X3 and Y 5.5 Coefficient β4 Null hypothesis: H0: β4=0 Alternative hypothesis: H1: β4≠0 The student t distribution 5% (2 sides) when n=39 ts=1,96 p-value=0.5120>0.05 => At 5% level of significance, we have enough evidence not to reject H => no linear relationship is recorded between X4 and Y 5.6 Hypothesis testing of R2 Hypothesis Using Fisher: R2=0.549005 → F =12.56453 According to the table p-value (F) = 2.22e – 06 < 0.05 * Conclusion, we reject hypothesis has statistical R2=0.596478: it means that regressors explain 59.65% of the variance of unemployment rate Conclusion: According to the results above, the coefficient β 2, β4 have no meaning in the model, β1, β3 have the statistical significance in the model Therefore, unemployment rate of Japan is affected by the following factors : GDP growth rate, inflation rate 13 We not eliminate variables X2,X4 because it doesn’t affect the estimated result VI Robustness check Multi-collinearity 1.1 VIF We can see from the chart that all the VIF values of the variables are less than 10 So, all the variables not show the collinearity problem 1.2 Correlation =0.5406 =0.3641 =-0.4003 =-0.5699 =-0.5413 =0-0.2387 So, Correlations between each couple of independent variables are low No multi-collinearity are found 14 Heteroskedasticity 2.1 Graph 2.2 White test H0: var (ui) = σ2 for all i H1: var (ui) ≠ σ2 for all i 15 The data table above shows that p-value = 0,369417 ≥ 0,05 => We have enough evidence to accept H0 Normality The hypotheses: H0: The sample data are not significantly different than normal H1: The sample data are significantly different than normal 16 P-value=0.9361>0.05 => We have enough evidence to accept H0 => Data follows normal distribution Auto correlation Null hypothesis: H0: No auto-correlation 17 p-value=1.38e-006 We have enough evidence to reject H0 VII Cure Multi-collinearity Model does not have multi-collinearity error Heteroskedasticity Model does not have heteroskedasticity error Normality The data follows normal distribution Auto-correlation Run model with Robust standard error 18 Auto-correlation test p-value=1.38e - 006