RELATIONSHIP BETWEEN INFLATION AND OTHER MACRO ECONOMICS FACTORS: COMPARATIVE STUDY OF GERMANY, JAPAN AND NEW ZEALAND

ABSTRACT


INTRODUCTION
The long-term objective of the economic strategy in every country is to achieve long-term economic growth while maintaining stable prices. As a result, fiscal policy aimed at increasing productivity and monetary policy aimed at maintaining price stability must be efficiently coordinated and implemented. Maintaining long-term economic expansion and price stability simultaneously can be difficult for legislators. Despite Keynesian theory, other economic notions emphasize that mild inflation is a growth stimulant (Mubarik, 2005). Increasing prices for products and services is the overall state of the economy, which is reflected in increased inflation (Sugihyanto, 2021). Because the buying power of individuals will decline as a result of rising prices for products and services, fewer people will be able to afford to buy the products and services that are no longer produced, which would reduce producer investment. Producers' investment will reflect economic growth if it declines, as will the nation's income. According to Ophanides and Solow (1990), there are three outcomes that CPI may have on output and growth: (i) none; (ii) positive; and (iii) negative. Wai (1959) and Bhatia are two studies that did not find any compelling scientific proof for either a favorable or adverse link between inflation and economic growth. Tobin (1965) presupposed that money might replace capital and invented the Tobin Effect, which is the idea that inflation positively affects growth. Fisher (1993), Barro (1995), Bruno and Easterly (1995), and many others discuss the Anti-Tobin Effect, which is the term used to describe the detrimental effect of inflation on development. The consumer price index is one of the most closely watched price statistics used by different governments in different countries. The consumer price index is used as an indicator of inflationary trends, and it is being observed because a slight impact on CPI may have a divergent impact on different variables. For example, the increase in CPI determines the rate at which different payments to Social Security recipients will rise each year. We will calculate the same impact between the consumer price index and the change in Agricultural Land, Urban Population, Trade, Military Expenditures, Primary Energy Consumption, Natural Gas Flaring basis, and oil on refining capacity basis. The main cause of our study today is to find the cointegration between these dependent and independent variables. The easiest definition of inflation is overall increases in the price of products and services within a country. For the past few years, inflation has been a problem faced by developed and underdeveloping economies. This study is conducted between the countries, namely Germany, Japan, and New Zealand, because, as we see below table for the past 20 years, these three countries also faced an inflation problem. In the case of Germany, according to www.aboutinflation.com, inflation was approximately 7% in 1974-75 while in the same years, according to www.aboutinflation.com inflation in Japan touch to the figure of 24% which is too high on the other hand, according to takeprofit.org, Newzeland also face this issue in recent few years. This study examines the relationship between Inflation dependent variables and Independent Variables are taken as Agricultural Land -%age of GDP (AL), Urban Population (UP), Trade-%age of GDP (T), Military Expenditure-%age of GDP (ME), Primary Energy Consumption (PEC), Natural Gas Flaring (NGF) and Oil -Refining Capacity (ORC).   The below section contains the studies that have already been conducted by various authors and have proven their stances through different techniques using econometrics. Previously the studies that have been conducted show a relationship between inflation and economic growth or within different variables holding the relationship between long-run and short-run phenomena. Oil prices, energy usage, and economic development have all had a substantial impact on Malaysian inflation rates discovered in the study by Talha et al. (2021). Adaramola and Dada (2020) analyze the relationship between inflation and macroeconomic factors and conclude that inflation and the real exchange rate have a considerable negative influence on economic growth, but interest rates and money supply have a positive impact on economic growth. Ahmad (2022) examines the relationship between inflation and economic growth; the findings of this study conclude that there is a negative and significant relationship between inflation and economic growth in Pakistan. BEDADA et al. (2020) examine the relationship between inflation and macroeconomic factors and conclude that there is a positive and significant relationship between CPI and money supply, real GDP, and total budget deficit. Milenković et al. (2020) examine the relationship between inflation and GDP, government expenditure, unemployment, and taxes and conclude that there is a positive and significant relation exists between inflation and said variables. Tien (2021) analyzes the relationship between inflation and GDP and concludes that a negative relationship exists between inflation and GDP. Al-Mutairi et al. (2020) examine the relationship between inflation and imports, GDP, Exchange rate, and money supply in Kuwait. The findings of this study conclude that there is a positive relation between inflation with imports and money supply, while there is a negative relation between GDP and exchange rate. Fisher (1993) has proven the relationship between inflation and economic growth. In this study, the data set consists of several macroeconomic variables, including consumer price index, Agricultural Land, Urban Population, Trade, Military Expenditure, Primary Energy Consumption, Natural Gas Flaring, and oil refinery capacity. Fisher argued that inflation distorts prices and effect the efficiency of the allocation of resources, and has a negative impact on economic growth. Barro (1997) has also shown a relationship between inflation and economic growth. Motley (1994) also included the inflation/consumer price index in his model to show the relationship between GDP and the consumer price index. Khan and Senhadji (2001) analyzed the relationship between the same variables in industrial and developing countries. They have used the techniques used by Chan and Tsay (1998) and Hansen (1999) with the help of new econometric techniques. Their results have shown that inflation rates above a specific level have a significant and negative effect on growth. There is also evidence that has supported the findings of Mundell (1963) and Tobin (1965) of a positive relationship between growth and inflation. Similar findings are also of Gosh & Philips (1998), with the relationship showing positive results once the inflation rate is less than 2-3%. Xiao (2009) proved that the said variables are positively related to the above three quarter's lags. He used the same multivariate cointegration method that we have used for this assignment. Christoffersen and Doyel (1998), on the other hand, have detected that below 13% inflation rate, there is no relationship between the two variables. The objective of this study is to find the relationship between CPI and other independent variables (Agricultural Land -%age of GDP (AL), Urban Population (UP), Trade-%age of GDP (T), Military Expenditure-%age of GDP (ME), Primary Energy Consumption (PEC), Natural Gas Flaring (NGF) and Oil -Refining Capacity (ORC)) in both long and short run using Johansen Multivariate Cointegration Technique.

METHODOLOGY Data source
Our study is based on three countries: Japan, New Zealand, and Germany. For our analysis in the current study, we will use annual data that covers 1980 to 2020 and collect all data from WDI and the British Petroleum website. CPI, agriculture land, urban population, Trade, and Military expenditure, are collected from WDI, and variables like Primary Energy Consumption, Natural Gas Flaring basis, and Oil on refinery capacity basis are taken from the British Petroleum statistical review of the World Energy Databank.

Method
We download all data from WDI and the British petroleum website and take that data in Excel after applying transformation approaches. Then, we take that data in E-views (find Correlation, VIF, descriptive table and following the lead of Chimobi (2010). We estimate the results by applying the Johansen cointegration method provided by Johansen (1988) and Johansen and Juselius (1990), along with Long run and short run results.
Here, Y= Inflation of Japan, New Zealand, and Germany = refers to each entity's unidentified intercept. AL = Agriculture Land (% of Land Area) of Japan, New Zealand, and Germany UP = Urban population (Total) of Japan, New Zealand, and Germany T = Trade (% of GDP) of Japan, New Zealand, and Germany ME = Military expenditure (% of GDP) Japan, New Zealand, and Germany PEC = Primary energy consumption (Exajoules) of Japan, New Zealand, and Germany NGF = Natural gas flaring (Billion cubic meters) of Japan, New Zealand, and Germany ORC = Oil refining capacity (Thousand barrels daily) of Japan, New Zealand, and Germany ℇ = refers to the error term The values are based on percentages of GDP and are on their constant value, which presidents the relative forms of the variables taken natural Logs taken on the values. We are using the entire variable in log form because when we use a variable in log form, then the unit of all variables becomes the same, and they are easily comparable with the rest of the world, and we easily interoperate them. Now we calculate the results of each country separately and, in the end, combine the results of all three as a summary. Table 1 presents variable names along with their log form version and proxy of Variables. Log transformed model is present below: Ln (CPI) = β 0 +β 1 Ln(AL) +β 2 Ln(UP) +β 3 (TRD) +β 4 Ln(ME) + β 5 Ln(PEC) +β 6 Ln(NGF) + β 7 (ORC) + ℇ (2) Bedada et al. (2020) also used this model style in their study. After discussing the descriptive statistic now, we are discussing VIF Tables 5, 6, and 7, which appear below. Table 5 shows the magnitude of variance inflation factors among the independent variables that we have taken for our study. The results show that where the VIF value is less than 10 by using the formula [1/1 -r2] the independent variables show no Multicollinearity between them. Tables 5 and 7 show that the VIF value of the LNME and LNUP in Germany and LNGF in New Zealand has Multicollinearity with a value of VIF greater than 10. As there are almost all the variables of all three countries have no Multicollinearity among them, so we go further to find the stationary of the variables by finding the trend.   Table 9, 10, and 11, which are attached below, and the criteria used to accept or reject the null hypothesis are present in Table 8. The data that we have picked is for 40 years , which shows that there is a trend in our data. We have applied two tests to find the unit root at the level and the first difference, Kwiatkowski-Phillips-Schmidt-Shin (KPSS) and NG Perron (NGP). At the level, the null hypothesis of NGP shows that the series is non-stationary, and the alternative hypothesis is stationary, whereas KPSS test results show us that at the level, the null hypothesis is the series is stationary, and the alternative hypothesis is non-stationary. The trend shows us that in the case of all three countries, Germany, Japan, and New Zealand, Ngperron results show that the natural log of CPI, natural log off AL, natural log of ME, natural log of ORC, natural log of PEC, natural log of T, and the natural log of NGF are non-stationery at the level because their calculated value correspondent to their critical value is greater at 10% significance level. Hence, we accept the null hypothesis that series is non-stationary, and KPSS also presents evidence of non-stationary in the case of all variables because their test value is greater than their critical value at a 1% significance level. Hence, we reject the null hypothesis that the series is stationary and accept the alternative hypothesis that the series is non-stationary. In the case of the First difference, all variables of Germany, Japan, and New Zealand are stationary. So in the case of all three countries, Germany, Japan, and New Zealand, our data series has the same order of integration, which is 1, and our data becomes spurious. To exclude spuriousness from the data, we apply Johansen and Juselius multivariate co-integration method. Table 9, 10, and 11 gives us values at the level and at the first difference of both KPSS and NG Perron tests.  Tables 12 and 13 reports the long-run cointegrating relation between CPI and their factors in the case of Germany, Japan, and New Zealand. The multivariate cointegration approach gives us two results: one in the form of a trace test and the other in the form of a maximum Eigenvalue test. The calculated value of our trace is found to be greater than 5% of its correspondence critical value, and we reject the null hypothesis and accept the alternative hypothesis and conclude that evidence of cointegration is present between dependent and independent variables up to the maximum value of six or at most 6 for Germany while eight or at most 6 for Japan and three or at most 3 for New Zealand, our maximum Eigenvalue test results shows that the calculated value of our Eigenvalue test is found to be greater than its correspondence critical value at 5% critical value and we reject the null hypothesis and accept alternative hypothesis and conclude that evidence of cointegration is present between dependent and independent variables at most 3 for Germany while at most 6 for Japan and at most 3 for New Zealand. This means that the natural log form of the dependent variable Consumer price indexes cointegrated with the natural log form of all other independent variables (Agricultural Land, Urban Population, Trade, Military Expenditure, Primary Energy Consumption, Natural Gas Flaring, and Oil Refining Capacity).  After discussing the Johansen Cointegration test. Further, to find out the cointegration between dependent and independent variables, we are going to calculate long-run and short-run variables. We calculate normalized and adjusted coefficients given in Table 14 and 15: if we talk about Germany, normalized coefficients show that Agricultural Land, Military Expenditure, Oil Refining Capacity, and Primary Energy Consumption has a positive and significant impact on the consumer price index of our Results are matched with Schurleet al. (2012), Starret al. (1984 and Arinze(2011), whereas Trade, Natural Gas Flaring, and Urban Population have a negative and significant impact on consumer price index, our results are similar with the studies of Joshi and Acharya (2010). This means that an increase in Agricultural Land or Military Expenditure or Oil Refining Capacity or Primary Energy Consumption by 1% will increase Inflation by 1.064, 0.050, .266, and 1.141, respectively. By increasing independent variables like Trade, Natural Gas Flaring, and Urban Population by 1%, they will decrease the consumer price index by.127, .259, and 1.77, respectively. The results also show that the impact on Oil Refining Capacity is the most among all variables upon the consumer price index. On the other hand, Japan normalized coefficients show that consumer price index, Agricultural Land, and Primary Energy Consumption have a negative and significant impact. Military Expenditure, Oil Refining Capacity, Trade, Natural Gas Flaring, and Urban Population positively and significantly impact the consumer price index. This means that an increase in Agricultural Land or Primary Energy Consumption by 1% will lead to a decrease in the consumer price index by 7.1902 and 0.7300, respectively. Whereas, by 1% increase in independent variables like Military Expenditure, Oil Refining Capacity, Trade, Natural Gas Flaring, and Urban Population will increase CPI by1.0319, 0.1258, 0.0725, 0.6081, and 1.0115, respectively. The result also shows that the impact on Agricultural Land is the most among all variables on the consumer price index. If we talk about New Zealand, the normalized coefficients show that the consumer price index, Oil Refining Capacity, Trade, and Urban Population have a negative and significant impact. In contrast, Military Expenditure, Primary Energy Consumption, and Natural Gas Flaring positively and significantly impact the consumer price index. Change in the consumer price index does not impact Agricultural Land in the longer run. This means that by 1% increase in Oil Refining Capacity, Trade, and Urban Population will decrease CPI by 1.7258, 4.2798and 1.6338, respectively. Whereas an increase in independent variables like Military Expenditure, Primary Energy Consumption, and Natural Gas Flaring by 1% will lead to an increase in CPI by 1.7309, 4.0811, and 2.4002, respectively. The result also shows that the impact on trade is the most among all variables upon the consumer price index in New Zealand. After discussing the long-run coefficients, we now discuss the short-run coefficients of Germany, Japan, and New Zealand based on Johansen multivariate co-integration method are present in Table 15. The results of the Germany-adjusted coefficients based on the multivariate cointegration method show that Military Expenditure and Urban Population have a significant and positive relationship with the consumer price index in the short run. Whereas primary consumption and Trade have a negative and significant impact on the consumer price index in the short run. This means that an increase in Military Expenditure and Urban Population by 1 percent will increase the consumer price index by 2.9806 and 0.1528, respectively. On the other hand, the increase in Primary Energy Consumption and Trade by 1% wills leads to a decrease in CPI by 1.445 and 3.654, respectively. Table 15 also shows us that Agricultural Land, Oil Refining Capacity, and Natural Gas Flaring have no impact, in the short run, on the consumer price index. The results of our first-period lag term error term are also found to be significant and negative, which confirms the existence of the convergence hypothesis for the model. The value of the coefficient (-.542) is negative and between -0.5 to -0.8. The disequilibrium resulted due to any macroeconomic shock will be removed by 54% every year and will return to stable and longterm equilibrium in just two years. On the other hand, according to Japan, the results of the adjusted coefficients based on the multivariate cointegration method show that only Oil Refining Capacity has a significant and positive relationship with the consumer price index in the short run. Whereas no variable has a negative and significant impact on the consumer price index in the short run. This means that an increase of 1% in Oil Refining Capacity will increase the consumer price index by 0.7449. Table  15 also shows us that Agricultural Land, Military Expenditure, Primary Energy Consumption, Trade, Natural Gas Flaring, and Urban Population has no impact, in the short run, on the consumer price index. The results of our first-period lag term error term are also significant and negative, confirming the existence of the convergence hypothesis for the model. The value of the coefficient (-0.0937) is negative and not between -0.5 to -0.8. The disequilibrium resulted due to any macroeconomic shock will be removed by 9.3% every year and will return to stable and longterm equilibrium in 10.75 years. While in the case of New Zealand the results of the adjusted coefficients based on the multivariate cointegration method show that only Oil Refining Capacity has a significant and positive relationship in the short run with the consumer price index. Whereas Primary Energy Consumption negatively and significantly impacts the consumer price index in the short run. Which means that by an increase in 1% in Oil Refining Capacity will increase consumer price index by 0.0941 and Primary Energy Consumption will decrease consumer price index by 0.0881. Table 15 also shows us that Agricultural Land, Military Expenditure, Trade, Natural Gas Flaring, and Urban Population have no impact, in the short run, on the consumer price index.
The results of our first-period lag term error term are also found to be significant and negative, which confirms the existence of the convergence hypothesis for the model. The value of the coefficient (-0.04015) is negative and not between -0.5 to -0.8. The disequilibrium resulted due to any macroeconomic shock will be removed by 4.0% every year and will return to stable and longterm equilibrium in 25 years. After discussing the long-run and short-run Cointegration now, in order to find the stability of our results, we apply error term diagnostic tests, and criteria for acceptance or rejection of the null hypothesis are present in Table 16. of all the tests shows us that the probability value of the tests is more than 0.1; therefore, the null hypothesis is accepted, and it is concluded that there is an absence of serial correlation in the model, the variance of the error term is homoscedastic, error term follows attributes of normal distribution and stability diagnostic CUSUM and CUSUM square in Figure 4 in case of all three countries blue line are lies within confidence interval its mean all countries show the stability of mean and variance of error term it also shows error term is not structurally unstable.

CONCLUSIONS
In this paper, we have investigated the relationship between consumer price and Agricultural Land, Urban Population, Trade, Military Expenditure, Primary Energy Consumption, Natural Gas Flaring, and Oil Refining Capacity using a time series of data from 1980 to 2020. In this study, we have applied Johnson's multivariate cointegration approach to estimating the long-run and short-run relationship between the variables. We have observed that although the same variables were used for the same time series data for all three different countries even still the results of them are quite different. The long-run coefficients that have a positive and significant impact on the consumer price index are different in Germany from New Zealand and Japan. Similarly, the short-run coefficients are different, having positive and negative significant impacts in the said countries. If we talk about Germany, normalized coefficients show that the consumer price index, Agricultural Land, Military Expenditure, Oil Refining Capacity, and Primary Energy Consumption have a positive and significant impact, Whereas Trade, Natural Gas Flaring, and Urban Population have a negative and significant impact on consumer price index. On the other hand, Japan normalized coefficients show that consumer price index, Agricultural Land, and Primary Energy Consumption have a negative and significant impact. Whereas Military Expenditure, Oil Refining Capacity, Trade, Natural Gas Flaring, and Urban Population have a positive and significant impact on the consumer price index. If we talk about New Zeland, their long run coefficient shows that the normalized coefficients show that consumer price index, Oil Refining Capacity, Trade, and Urban Population have a negative and significant impact, Whereas Military Expenditure, Primary Energy Consumption, and Natural Gas Flaring has a positive and significant impact on consumer price index. Change in the consumer price index does not impact Agricultural Land in the longer run. We also apply diagnostic tests; all the diagnostic lm serial correlation, Heteroskedasticity, and normality, show probability values greater than 0.1, so we conclude that our variables are not serially correlated, not abnormally distributed, not heteroscedastic, and functional form is not miss specified while CUSUM and CUSUM square shows error term is not structurally unstable. Therefore, the study recommends that governments of the aboveselected countries should take more initiatives to increase their urban population and trade because these activities help decrease inflation in New Zealand, Germany, and Japan. On the other hand, decrease their focus on military expenditure and primary energy consumption because these two variables take part in the acceleration of inflation within a country.