Analysis of the Effect of Unemployment Inflation and Population on Poverty (Data Study of North Sumatra Province)
Nasution AR, Arismunandar MF, Muda I, Soemitra A and Sugianto S
Published on: 20240803
Abstract
This study aims to analyze the effect of open unemployment, inflation, and population on poverty in North Sumatra. The data used in this study is secondary data obtained from the Central Bureau of Statistics (BPS) and other relevant sources during a certain period. The analysis method used is multiple linear regression to identify the relationship between the independent variables (open unemployment, inflation, and population) and the dependent variable (poverty). The results show that open unemployment has a positive significant effect on poverty, which means that an increase in the unemployment rate will increase the poverty rate. Inflation was also found to have a significant influence on poverty, where an increase in inflation tends to increase the poverty rate. In addition, population has a positive significant effect on poverty, which suggests that population growth that is not matched by an increase in employment opportunities and income can increase the poverty rate. These findings provide important implications for policymakers in North Sumatra to focus on strategies to reduce unemployment, control inflation, and manage population growth in an effort to reduce poverty in the region.
Keywords
Open unemployment; Inflation; Population; Poverty; North SumatraIntroduction
Globally, economic development aims to achieve fairly high economic growth, realizing a prosperous, just and prosperous life for the people and the country. This is in accordance with the content of the ideals of this nation to create a just and prosperous society through economic development and growth as one of the efforts to overcome poverty. Because poverty is a problem that never ends and all countries must experience this problem either at a small level or at a concerning level. Poverty is not a problem that can be left unaddressed; it must be addressed at its root. If not addressed, it will become an obstacle to regional development and equity. Individuals are said to be poor if they are unable to utilize the choices and opportunities available to them to meet their basic needs, such as health care, a decent standard of living, a sense of freedom, independence, selfreliance, appreciation, and respect from others. (World Bank, 2015). High poverty will cause a very detrimental impact on a country. Not only social impacts, but it directly impacts individuals and impacts the wider community. In North Sumatra Province, poverty is an extremely crucial problem, not only is its condition increasing, but the consequences of poverty do not only cover the economic scope. It also causes social problems and makes it difficult for people to fulfill their needs, resulting in actions such as theft and increased crime rates. The population of North Sumatra also continues to increase every year, which if not balanced with an increase in human resources will make poverty even more difficult to overcome. In addition to population, inflation in this province is increasing every year, causing the price of raw materials to continue to increase and can trigger poverty to rise.
Poverty is also influenced by high inflation and a country's population that is not matched by the quality of its human resources (Nelson and Liebstein) the relationship between population growth and the level of community welfare has an influence. According to Nelson and Leibstein, rapid population growth in developing countries does not significantly increase the level of public welfare and in the long run will reduce welfare and increase the number of poor people. The increase in population does have a positive effect, namely the availability of abundant human resources and the number of entrepreneurs that arise, but a large population is also closely related to poverty. This happens because the production of agricultural products is slower than human growth itself, thus it will have an effect on the difficulty of humans to fulfill their needs, then it will have an impact on the increasing poverty rate. Another factor that causes poverty is inflation. One economic indicator that is closely related to poverty is inflation. According to Mankiw [1] if inflation rises and the value of the real currency fluctuates significantly, higher inflation will certainly raise the poverty line. Inflation itself has both good and bad impacts, the bad impact of inflation is that it will increase the price of goods and services in general continuously which will directly increase the poverty rate. At first, people were able to fulfill their needs, but because of the higher inflation, it will be more difficult for people to fulfill their own needs. The inflation rate in Indonesia itself fluctuates from year to year. Another influential factor that causes the poverty rate to increase is the high unemployment rate in an area. Losing a job is one of the most difficult events in a person's life. According to Sukirn [2], the negative impact of unemployment is to reduce people's income, which in turn reduces the level of prosperity of an individual. The opportunity for people to be trapped in poverty because they have no income will certainly increase due to the decrease in welfare benefits due to unemployment. Many people, especially in developing countries, rely solely on their livelihoods to survive. Losing a job will leave a person without income and if they do not get a job as soon as possible, it will result in not being able to fulfill their needs and losing their selfesteem. Unemployment itself is often used as an issue by politicians to garner votes by promising policies and proposals to create new jobs. Based on data published by the Central Bureau of Statistics (BPS), the number of poor people in North Sumatra from 20032019 has decreased by 7.06%. However, in 20192021 there was a covid 19 pandemic crisis which caused an increase in the number of poor people by 0.18% in 2021. The decline in the poor population from 2006 to 2019 before the pandemic period and it can be said that the government's efforts to reduce poverty are quite good. After 2021 there was a decrease in the poor population due to government efforts to restore stability to the country's economic life. According to the explanation of BPS from 20032023, there was a decrease in the number of poor people in North Sumatra, in 2013 there was an increase in fuel prices but in 20132014 there was a decrease in the poor population from 10.39% to 9.85% in 2014. From these data, it can be said that the number of poor people after the increase in fuel prices, which caused the inflation rate to rise, did not have an impact on the increase in the number of poor people in 20132014. According to Nelson and Liebstein (1983) a country's rapid population growth will not result in a significant increase in the level of social welfare, which in turn will result in a decrease in welfare and an increase in the number of poor people. The population of North Sumatra published by the Central Bureau of Statistics continues to experience a significant increase every year. In 2003, the population of North Sumatra was 11890399 people and continued to increase to 15386640 people in 2023. If reviewed more deeply, the government has actually carried out several strategies to control the population, one of which is the family planning program (KB), but it can be seen as data from the Central Statistics Agency of Indonesia (BPS) states that the population in North Sumatra continues to increase. BPS data states that the population will continue to increase with fluctuating poverty, this large quantity of resources should be utilized by the government as a factor of production in an effort to increase economic growth and the government can balance the quantity of employment opportunities.
Furthermore, what affects the poor is inflation, according to Mankiw [1] if inflation rises and the value of the real currency fluctuates significantly, higher inflation will certainly raise the poverty line Inflation that increases will result in an increase in the number of poor people, especially if the purchasing power and income of lowincome groups are not accompanied by an increase. BPS data shows that the percentage of inflation in North Sumatra experienced fluctuating movements. In 2005 inflation was at its highest point at 22.41%, which was the first year of SBY's cabinet government, the policy of 100 percent fuel price increase which resulted in a sharp increase in the price of daily necessities including all existing goods and services. However, in the following year, the inflation growth rate returned to a safer level and was adjusted. However, in 2008 there was another increase due to the rise in international crude oil prices which was triggered by a comprehensive increase in fuel prices. The economy of North Sumatra and the nation as a whole was severely affected by this condition. Until 2012 to 3.67%, one of the efforts made by the government in 2012 to reduce inflation was to strengthen food security by implementing a warehouse receipt system (RSG) and increasing food production. Then again experienced a fairly high increase in inflation due to the increase in fuel in 2013 to reach 10.18%. Until 2019, it experienced a significant decline of 2.33% because prices had begun to adjust to the fuel increase. In 2020, it experienced a pandemic period but inflation experienced an increase and even decreased by 1.96%, then in 2021, this year, it is still experiencing a pandemic period, then inflation rises to 6.14%. From 2021 to 2023, inflation has decreased by 2.94%. This is one of the results of the government's efforts to restore economic stability. Various policies are carried out to reduce the inflation rate, especially in North Sumatra. The above things must be considered to see the poverty rate has decreased significantly and is proportional to the number of people who increase every year.
In addition to population and inflation, unemployment also affects the poverty rate. According to Sukirno [2], the negative impact of unemployment is to reduce community income, which in turn reduces community income, which in turn reduces the level of prosperity of an individual. The chance for people to be trapped in poverty because they have no income will certainly increase due to the decrease in welfare benefits due to unemployment.
According to BPS data, North Sumatra's open unemployment rate experienced fluctuating movements. The highest phase of North Sumatra's open unemployment rate was in 2004 at 13.75%. Based on data published by the Central Bureau of Statistics of North Sumatra, the unemployment rate was high in 3 districts in North Sumatra, namely Central Tapanuli, Labuhan Batu and Simalungun. However, in the following year there was a decrease. In 2005 the open unemployment rate was 10.98% due to a natural event, namely the earthquake in Nias, which caused a significant spike in the unemployment rate in 2005 and has decreased from year to year, this is inseparable from the domestic apprenticeship program and the success of the North Sumatra government in reducing the open unemployment rate. In 2020 and 2021 North Sumatra experienced a covid pandemic which caused the open unemployment rate to increase again. So that the factors that affect poverty in this case that want to be studied are population, unemployment and inflation. Several previous studies have examined related to this research. According to Mahsunah [3], concluded that population has no effect on poverty while education and unemployment affect poverty. Furthermore, according to Suharianto and Lubis [4], concluded that simultaneously and partially unemployment and inflation have a positive and significant effect on poverty in North Sumatra Province.
Theoretical Foundation
 overty
Poverty according to Maipita [5] is a phenomenon, a reality that has never been forgotten from the face of this earth. Poverty arises due to differences in abilities, opportunities and differences in resources. Meanwhile, according to BPS (2012), poverty is the inability of individuals to meet the needs of a decent life. According to Paul Spicker Wijayanto [6], the causes of poverty can be divided into four madzab, namely individual explanation, familial explanation, subcultural explanation, structural explanation. In addition, poverty is a condition in which a person is below the poverty line, also known as the standard value line of minimum needs for both food and nonfood. Or sometimes referred to as the poverty threshold. The poverty trap cycle theory is a series of events that influence each other in such a way that can lead to a situation where a country will remain poor and will experience many difficulties in achieving a more prosperous level of development than before. Characterized by underdevelopment, market imperfections, and the absence of capital that can lead to low productivity. The result of low productivity itself is the low income they get. Low income will lead to low levels of investment and investment, both in the form of human investment and capital investment. Low levels of investment can result in underdevelopment and other related matters. According to Kuncoro [7], the factors that affect poverty are:
 At a macro level, poverty arises as a result of unequal ownership of limited resources, which results in an unequal distribution of income; the poor have only a limited amount of resources and their quality is low.
 Poverty arises due to the low quality of the resources of the poor. There are differences in knowledge, skills, and this causes inequality, individuals who have extensive knowledge will have high creativity so as to produce maximum productivity and wages.
 Poverty arises due to differences in access and capital. This difference in access and capital causes an individual to feel worried about having limited capital when they want to start a business. On the other hand, some groups of people with access to existing capital sources will find it much easier to earn maximum wages.
 Total Population
The Indonesian Central Bureau of Statistics [17] states that "residents are all people who live in the Geographical Area of the Republic of Indonesia for more than 6 months they or whose stay is less than 6 months, but have the aim of settling. On the other hand, according to Said [18], it states that the population is the average "number of inhabitants of an area at a certain time and the occurrence of the results of the demographic process, namely birth rates, death rates, immigration". According to [8] there are three factors that affect population size, namely:
 Births (Fertility)
Fertility is the ability of a woman to give birth. Fertility is related to population births which involves the number of babies born alive. Fertility or birth is one of the factors that contribute to population growth.
 Mortality
Mortality is the death of an individual. Mortality is closely related to the death rate of the population in a region. Not all deaths are recorded in demography, for example: the number of miscarriages and still births are not counted as deaths. A high or low mortality rate in an area not only impacts the rate of population growth there, but also serves as a measure of the health and wellbeing of the population there.
 Population Movement (Migration)
Migration is the movement of individuals from one region or area to another, either temporarily or permanently. There are two types of migration, namely:
1) National migration is a movement of people from one region to another but still within one country.
2) International migration is the movement of people from one country to another.
According to Nelson and Leibstein [19], there is a relationship between population growth and the level of public welfare. Nelson and Leibstein point out that rapid population growth in developing countries causes the level of public welfare to not experience a significant increase and in the long run will decrease welfare and increase the number of poor people.
In accordance with an empirical study by Whisnu Adhi [20], where the effect of population variables on poverty was investigated. Based on the research findings, there is a positive correlation between population and poverty, meaning that the number of poor people increases as the population increases.
 Unemployment
A person who does not work at all, is looking for work, works less than two days a week, or is trying to find a decent job is known as unemployed or unemployed [9]. Someone who works but has not been able to get a permanent job is said to be unemployed [10]. Meanwhile, according to BPS (2012) someone in the labor force (over the age of 15) who is looking for work but is unsuccessful is considered unemployed. According to Sukirno [11], the kinds of unemployment based on working hours can be classified as hidden unemployment, seasonal unemployment, underemployment, and unemployment. People's chances of being trapped in poverty due to lack of income will certainly increase as welfare benefits fall due to unemployment. If there is a lot of unemployment in a country, political and social unrest will always be rampant, which will harm the welfare of the people and their economic growth opportunities in the long run. According to Yulistiyono [21], the factors that cause unemployment are as follows:
 The size of the labor force is not balanced with employment opportunities. An imbalance can occur when the size of the labor force is greater than the number of available job opportunities, but the opposite is very rare.
 Unbalanced employment structure the imbalance between the required employment structure and the background of workers seeking employment causes unemployment.
 The need for the number and type of educated workers and the supply of educated workers are not balanced. If the number of job opportunities is the same or greater than the labor force, unemployment does not necessarily not occur because there is not necessarily a mismatch between the level of education needed and available.
 Increasing the role and aspirations of women workers in Indonesia's overall labor force structure.
 The supply and utilization of labor between regions is not balanced. The number of labor force in a region may be greater than employment opportunities while in other regions the opposite situation may occur, resulting in the movement of labor from one region to another and even from one country to another.
 Layoffs due to the economic crisis in a region or country.
According to Sukirno [2], the negative impact of unemployment is to reduce people's income, which in turn reduces the level of prosperity of an individual. The opportunity for people to be trapped in poverty because they have no income will certainly increase due to the decrease in welfare benefits due to unemployment [12]. When there is a lot of unemployment in a country, political and social unrest will always occur, which is detrimental to the welfare of the people and their economic growth opportunities in the long run.
Community income will decrease due to unemployment, which will have an impact on the low human development index because people will not be able to fulfill their basic daily needs such as education and health. If unemployment continues on a protracted basis, it will have an impact on the economy of a family, which will lead to poverty. If a poor family is too large, it will definitely affect the way an individual's social behavior. The pressure to make ends meet can also change a person's psychological pattern, allowing them to do things that are considered abnormal and lead to violence and crime. Criminality will inevitably cripple economic activities, causing a loss of morale among the victims. The result of protracted unemployment is a cycle of poverty that leads to high crime rates. Therefore, to break this cycle, unemployment, poverty and crime must be addressed simultaneously. Unemployment must be ended immediately, and various parties, especially the government, play an important role in this [22].
 Inflation
Inflation is the process of general price increases in the economy. On the other hand, according to Mandala Manurung [13], the concept of inflation is a general and continuous increase in commodity prices. According to Adi Warmankarim, inflation generally means that the price level of goods or services generally increases within a certain period of time. Inflation can be viewed as a monetary phenomenon caused by the depreciation or reduction in the value of a unit of monetary calculation of a commodity. Then Inflation is also the tendency of commodity prices to continue to increase. The increase in commodity prices is comprehensive, not only for some types of goods. It cannot be said to be inflation if the price increase is only for a few goods, but if the increase in an item can affect other goods, it can only be called inflation [14]. Meanwhile, according to Sukirno [2]. States that high inflation causes the fixed (real) income of the community to decrease in the form of cash or money deposits in banks. According to Santosa [15], the factors that cause inflation are based on Keynes Theory and Structuralist Theory. The Keynes Theory explains the factors that cause inflation to occur because a society wants to live beyond the limits of its economic capacity (disposable income). This is translated into a condition where people's demand for goods exceeds the amount of goods available, so an inflationary gap appears. This inflationary gap arises because people manage to translate their aspirations into effective demand for goods [16]. Inflation will continue as long as the effective demand of the society exceeds the amount of output that can be produced by the society. Inflation will only stop when the total effective demand does not exceed the prevailing prices of available output. In terms of money supply, high growth is often the cause of high inflation rates. The increase in money supply will lead to an increase in aggregate demand. If this condition is not matched by growth in the real sector, it will cause prices to rise (inflation). According to Mankiw [1], if inflation rises and the real value of the currency fluctuates significantly, higher inflation will certainly raise the poverty line. Higher inflation will lead to an increase in the number of poor people, especially if the purchasing power and income of lowincome groups do not increase [23].
Discussion and Result
This study uses the Error Corection Model (ECM) to determine shortterm and longterm relationships. In this method there are several stages of testing, including testing data stationarity and testing the estimation results [24].
 Root Test
Before conducting the ECM test, the steps taken are testing the estimation results by testing the data using the unit root test to analyze whether the time series data is stationary. Stationarity is an important requirement in processing time series data. The data stationarity test was conducted using the Augmented Dickey Fuller (ADFTest) method at the level level, and the results are listed in Table 4.1. In the ADF test, the variables of population, open unemployment, and inflation show probability values greater than α = 5% (0.05) at the level or I (0). Thus, it can be concluded that all these variables are not stationary at the level. Therefore, it is necessary to test the degree of integration or unit root test at the first difference level to determine at what degree the data will become stationary, with the aim of avoiding correlation.
Table 1: Root Test.
Null Hypothesis: Unit root (individual unit root process) 

Series: X1, X2, X3, Y 

Date: 06/29/24 Time: 13:50 

Sample: 2003 2023 

Exogenous variables: Individual effects 

Automatic selection of maximum lags 

Automatic lag length selection based on SIC: 0 

Total (balanced) observations: 80 

Crosssections included: 4 

Method 
Statistic 
Prob.** 
ADF  Fisher Chisquare 
12.8412 
0.1174 
ADF  Choi Zstat 
0.60923 
0.2712 
 Degree of Integration Test
The results of the degree of integration test at the first difference level are listed in Table 2. The results show that the probability values of the population, open unemployment, and inflation variables are lower than α = 5% (0.05), indicating that the data tested are stationary at the first difference level [25].
Table 2: Degree of Integration Test Result.
Null Hypothesis: Unit root (individual unit root process) 

Series: X1, X2, X3, Y 

Date: 06/29/24 Time: 13:56 

Sample: 2003 2023 

Exogenous variables: Individual effects 

Automatic selection of maximum lags 

Automatic lag length selection based on SIC: 0 to 3 

Total number of observations: 73 

Crosssections included: 4 

Method 
Statistic 
Prob.** 
ADF  Fisher Chisquare 
122.478 
0 
ADF  Choi Zstat 
8.13483 
0 
 Cointegration Test
After conducting the stationarity test, the next step is to conduct a cointegration test to determine whether there is cointegration in the variable data that shows the short term and longterm relationship between these variables. The method used in the cointegration test is the Engle Granger Method. The results of the cointegration test using the EngleGranger Method are in Table 3.
Table 3: Data Test Result.
Null Hypothesis: D(ECT) has a unit root 

Exogenous: Constant 

Lag Length: 2 (Automatic  based on SIC, maxlag=4) 


tStatistic 
Prob.* 

Augmented DickeyFuller test statistic 
3.1839462 
0.0390917 

Test critical values: 
1% level 
3.8867513 

5% level 
3.052169 

10% level 
2.6665932 
In this study, if the stationary residuals at the level level show a significant tstatistic value at the 1% significance level and (Prob. 0.039) which is stationary at alpha α =1%, this indicates that the data has cointegration. With cointegration, it can be concluded that there is a shortterm relationship and a longterm relationship between these variables. With the steps that have been taken and all steps have met the requirements, the next step is to conduct an ECM (Error Correction Model) regression analysis [26].
 ShortTerm Error Correction Model (Ecm)
 Ecm Test Result
Based on the cointegration results that have been done previously, the variables that affect poverty, namely Inflation, Population and Open Unemployment, have a cointegration relationship in the first difference, so we proceed to the ECM modeling stage.
Table 4: Data Test Result.
Dependent Variable: D(Y) 

Method: Least Squares 

Date: 06/29/24 Time: 14:25 

Sample (adjusted): 2004 2023 

Included observations: 20 after adjustments 

Variable 
Coefficient 
Std. Error 
tStatistic 
Prob. 
C 
0.436861 
0.272793 
1.601436 
0.1301 
D(X1) 
0.131377 
0.088387 
1.486386 
0.1579 
D(X2) 
0.029241 
0.021721 
1.346198 
0.1982 
D(X3) 
2.54E07 
1.38E06 
0.183756 
0.8567 
ECT(1) 
0.667359 
0.207488 
3.216372 
0.0058 
Rsquared 
0.440424 
Mean dependent var 
0.387 

Adjusted Rsquared 
0.291204 
S.D. dependent var 
0.666113 

S.E. of regression 
0.560801 
Akaike info criterion 
1.893415 

Sum squared resid 
4.717459 
Schwarz criterion 
2.142348 

Log likelihood 
13.93415 
HannanQuinn criter. 
1.942009 

Fstatistic 
2.951505 
DurbinWatson stat 
1.626931 

Prob(Fstatistic) 
0.055301 

Based on the estimation results of the Error Correction Model above, the shortterm ECM equation results are as follows: DPE = 0.436 + 0.1313 Population  0.029 Inflation + 2.54 Open Unemployment 0.667ECT
Based on the estimation results above, it can show that:
1) The constant value of 0.436 means that if the variables of Total Unemployment (X1), Inflation (X2), and Population (X3) are equal to zero, then Poverty (Y) is 43.6 percent.
2) R2 = 0.44 which means that the variables of Population, Inflation and Open Unemployment are able to explain the effect of Poverty by 44% and the remaining 54% is explained by other variables not included in the estimation model [27].
3) Based on the results of the simultaneous test conducted to see the significance together, the probability value (FStatistic) is 0.055. This value is greater than the significance level α = 5% (0.05), which means that together the dependent variables do not have a significant influence on the independent variables with a confidence level of 95%.
4) The ECM (Error Correction Model) model includes the ECT (Error Correction Term) variable. The ECT variable regression coefficient is an adjustment coefficient which is the speed of adjustment between the actual value and the desired value that will be eliminated in one period. Poverty variables are not only influenced by Population, Inflation and Open Unemployment alone but also influenced by the ECT error term variable. The ECT coefficient in this study is 0.667, which means that the difference between what affects Poverty and its equilibrium value is equal to 0, 66. The results of shortterm regression or ECM, obtained ECT probability value of 0.005 < α = 5% (0.05) which means significant. The ECT coefficient value must be negative and significant, so it can be said that the ECM model used is appropriate.
Classical Assumption Test of ShortTerm Model
 Normality Test
In this study, the normality test was carried out using the JarqueBerra test (JB Test) with a significance level of 1%, 5%, and 10%.
Figure 1: Normality Test Result.
Based on the results of the normality test on the shortterm equation, a probability value of 0.84 is obtained. This value is greater than the significance level α = 5%. Therefore, it can be concluded that the data used in the shortterm regression ECM model has a normal distribution [28].
2) Heteroscedasticity Test
Table 5: Heteroscedasticity Test Result.
Heteroskedasticity Test: White Null hypothesis: Homoskedasticity 

Fstatistic 
6.645731 
Prob. F(14,5) 
0.0235 
Obs*Rsquared 
18.98001 
Prob. ChiSquare(14) 
0.1657 
Scaled explained SS 
9.527463 
Prob. ChiSquare(14) 
0.7959 
Based on the results of data processing in the heteroscedasticity test obtained chi square probability of Obs * RSquared of 0.1657 where the value is greater than α = 5%, it can be said that in the shortterm equation model there is no heteroscedasticity problem [29].
3) Multicollinearity Test
Table 6: Data Test Result.

Coefficient 
Uncentered 
Centered 
Variable 
Variance 
VIF 
VIF 
C 
0.074416 
4.732384 
NA 
D(X1) 
0.007812 
1.341257 
1.337143 
D(X2) 
0.000472 
1.145414 
1.145334 
D(X3) 
1.91E12 
4.905526 
1.201647 
ECT(1) 
0.043051 
1.433118 
1.431551 
Based on the results of the Multicollinearity Test data processing from shortterm calculations, the VIF variable value is <10, it is concluded that the data is free from multicollinearity symptoms.
4) Autocorrelation Test
Table 7: Autocorrelation Test Result.
BreuschGodfrey Serial Correlation LM Test: 

Null hypothesis: No serial correlation at up to 2 lags 

Fstatistic 
1.617069 
Prob. F(2,13) 
0.236 
Obs*Rsquared 
3.984368 
Prob. ChiSquare(2) 
0.1364 
From the calculation of the shortterm equation, it is found that the probability value of Obs* RSquared is 0.326. This value is greater than the significance level of α = 5%, which indicates that there are no autocorrelation symptoms in the short term equation with the ECM model [30].
 LongTerm Error Connection Model
 Ecm Test Result
Table 8: ECM Result.
Variable 
Coefficient 
Std. Error 
tStatistic 
Prob. 
C 
32.03278 
4.313181 
7.426717 
0 
X1 
0.346186 
0.118285 
2.926722 
0.0094 
X2 
0.01267 
0.044381 
0.285484 
0.7787 
X3 
1.71E06 
2.66E07 
6.445018 
0 
Rsquared 
0.910414 
Mean dependent var 
11.19762 

Adjusted Rsquared 
0.894605 
S.D. dependent var 
2.443528 

S.E. of regression 
0.793282 
Akaike info criterion 
2.544367 

Sum squared resid 
10.69803 
Schwarz criterion 
2.743324 

Log likelihood 
22.71585 
HannanQuinn criter. 
2.587546 

Fstatistic 
57.58739 
DurbinWatson stat 
1.31905 

Prob(Fstatistic) 
0 

Based on the estimation results of the Error Correction Model above, the long term ECM equation results are as follows:
KMt = 32.032+ 0.346 Total Populationt 0.012 Inflationt  1.71 Unemploymentt + ????t
1) The constant value of 32,032 means that if the variables of Total Population (X1), Inflation (X2), and Open Unemployment (X3) are equal to zero, then Poverty (Y) is 32 percent.
2) R2 = 0.9104 which means that the variables of Population, Inflation and Open Unemployment are able to explain the effect of Poverty by 91% and the remaining 9% is explained by other variables not included in the estimation model.
3) Based on the results of the simultaneous test conducted to see the significance together, an estimate with a probability value (FStatistic) of 0.00000 was obtained. This value is smaller than the significance level α = 5% (0.05), which indicates that together the dependent variables have a significant influence on the independent variables with a confidence level of 95% [31].
 Classical Assumption Test of LongTerm Model
1) Normality Test
In this study, the normality test was carried out using the JarqueBerra test (JB Test) with a significance level of 1%, 5%, and 10%.
Figure 2: Normality Test Result.
Based on the results of the normality test on the shortterm equation, a probability value of 0.58 is obtained. This value is greater than the significance level α = 5%. Therefore, it can be concluded that the data used in the longterm regression ECM model has a normal distribution [32].
Heteroscedasticity Test
Table 9: Heteroscedasticity Test Result.
Fstatistic 
4.976746 
Prob. F(9,11) 
0.0076 
Obs*Rsquared 
16.85953 
Prob. ChiSquare(9) 
0.051 
Scaled explained SS 
8.735269 
Prob. ChiSquare(9) 
0.4621 
Based on the results of data processing in the heteroscedasticity test, the chisquare probability of Obs * RSquared is 0.051 where the value is greater than α = 5%, it can be said that in the longterm equation model there is no heteroscedasticity problem [33].
3)Multicollinearity Test
Table 10: Multicollinearity Test Result.
Variable 
Coefficient 
Std. Error 
tStatistic 
Prob. 
C 
2.373932 
4.877125 
0.486748 
0.6335 
X1 
0.12471 
0.158552 
0.786565 
0.4438 
X2 
0.040197 
0.059892 
0.671153 
0.5123 
X3 
1.22E07 
2.92E07 
0.417967 
0.6819 
RESID(1) 
0.439873 
0.365855 
1.202317 
0.2479 
RESID(2) 
0.07331 
0.293784 
0.249519 
0.8063 
Based on the results of the Multicollinearity Test data processing from longterm calculations, the VIF variable value is <10, it is concluded that the data is free from multicollinearity symptoms.
4)Autocorrelation Test
Table 11: Autocorrelation Test Result.
BreuschGodfrey Serial Correlation LM Test: 

Null hypothesis: No serial correlation at up to 2 lags 



Fstatistic 
0.734496 
Prob. F(2,15) 
0.4962 
Obs*Rsquared 
1.873146 
Prob. ChiSquare(2) 
0.392 
From the calculation of the shortterm equation, it is found that the probability value of Obs* RSquared is 0.392. This value is greater than the significance level of α = 5%, which indicates that there are no autocorrelation symptoms in the longterm equation with the ECM model [34].
 Hypothesis Test
 T Test (Partial)
The t test is an individual test conducted to determine the effect of the independent variable on the dependent variable. Based on the calculation results, df = (nk) is obtained, df = (214) = 17, where the ttable value is 2.11991. Then the partial test calculation for the short term is:
1) Open Unemployment variable, tcount = 1.4863 < ttable = 2.11991 and prob value of 0.157 > 0.05, so it can be concluded that Ho is accepted Ha is rejected, meaning that there is no effect of inflation on poverty in the short term in North Sumatra.
2) Inflation variable, the value of tcount = 1.346 < ttable = 2.11991 and the prob value of 0.198 > 0.05, so it can be concluded that Ho is accepted Ha is rejected, meaning that there is no effect of population size on poverty in the short term in North Sumatra.
3) Population variable, tcount value = 0.1837 < ttable = 2.11991 and prob value of 0.856 > 0.05, so it can be concluded that Ho is accepted Ha is rejected, meaning that there is no effect of unemployment on the amount of poverty in the short term in North Sumatra.
Then for the Longterm partial test namely:
1)Open Unemployment variable, the value of tcount = 2.926 > ttable = 2.11991 and the prob value of 0.000 <0.05, so it can be concluded that Ho is rejected Ha is accepted, meaning that there is an effect of inflation on poverty in the long run in North Sumatra.
2)Inflation variable, tcount value = 0.285 < ttable = 2.11991 and prob value of 0.778 > 0.05, so it can be concluded that Ho is accepted Ha is rejected, meaning that there is no effect of population size on poverty in the long run in North Sumatra.
3)Population variable, the value of tcount = 6.44 < ttable = 2.11991 and the prob value of 0.000 < 0.05, so it can be concluded that Ho is rejected Ha is accepted, meaning that there is an effect of unemployment on the amount of poverty in the long run in North Sumatra.
 F Test (Simultaneous)
In the shortterm analysis, the simultaneous test results using the Ftest show that the probability of Fcount is 0.055 > α = 5% (0.05). Thus, it can be concluded that inflation, population, and open unemployment together or simultaneously have no significant influence on the poverty rate in North Sumatra. That is, based on the shortterm estimation results, the independent variables in the model together do not have a significant effect on the dependent variable.
In the longterm analysis, the simultaneous test results using the Ftest show that the Fcount probability is 0.00000 < α = 5% (0.05). Thus, it can be concluded that inflation, population, and open unemployment together or simultaneously have a significant influence on the poverty rate in North Sumatra. That is, based on the longterm estimation results, the independent variables in the model jointly have a significant effect on the dependent variable.
 Results of the Determinant Coefficient (R2)
Based on the results of the Error Correction Model (ECM) analysis for the short term, the coefficient of determination (R2) is 0.44. This indicates that 44% of the variation in the dependent variable can be explained by variations in the independent variable, while the remaining 56% is explained by other factors outside the model. Based on the longterm analysis results, the coefficient of determination (R2) is 0.910. This indicates that 91% of the variation in the dependent variable can be explained by variations in the independent variable, while the remaining 9% is explained by other factors outside the model [35].
Conclusion
Even with the same data, if used for different functions, it will produce different types of research results. By using the Error Correction Model, it is possible to analyze the short and long term. This can help the government as advice in decision making. The above modeling shows different results in shortterm and longterm research. Furthermore, this research model can be used in various types of quantitative research, especially with time series data that can produce different decisions so that in making decisions not only think about the short term, but also longterm plans. Inflation in the short and long term has no effect on poverty in North Sumatra, therefore it is expected that the local government can maintain and continue to suppress the inflation rate to remain stable. To encourage inflation to remain stable, it can do several ways, in the short term the government itself can make cheap market operations in traditional markets, especially on big days and tighten security and supervision so that there are no irresponsible people who commit acts of hoarding goods or important commodities before the big day. In the long term, the government is expected to guarantee distribution channels in the form of good roads to toll roads so that shipments are more efficient and reduce shipping costs aimed at stabilizing prices. The population in the short term has no effect on poverty, but in the long term it affects poverty, so it is hoped that the government needs to strengthen human resources in order to improve and improve the skills, knowledge or quality of the workforce itself. Strengthening the quality of the workforce itself at this time needs to be done in order to meet the standard provisions of the industry or company and is expected to have competent human resources. Apart from increasing training, it is hoped that the central and regional governments can further control the population by encouraging more such as the disaster family program so that there is no overpopulation which can lead to various problems both unemployment problems and a decrease in the quality of human resources which can be the main problem of poverty. Open unemployment in the short term has no effect on poverty and in the long term open unemployment affects poverty in North Sumatra, therefore it is hoped that the government can make policies that encourage investors to be more interested in investing, especially in the North Sumatra region itself so that new jobs can be created, an example of a problem that often occurs, especially in the Medan City area, is the problem of rampant extortion or illegal parking which is often highlighted on various social media, as a result of this also makes investors less interested in investing in Medan City.
References
 Mankiw NG. Principles of economics. SouthWestern College Pub. 2003.
 Sukirno S. Makro Ekonomi Teori Pengantar. Jakarta PT Rajawali Pers. 2016.
 Mahsunah D. Analisis Pengaruh Jumlah Penduduk, Pendidikan Dan Pengangguran Terhadap Kemiskinan Di Jawa Timur. Jurnal Pendidikan Ekonomi (JUPE). 2013; 1: 117.
 Suharianto J and Lubis HR. Pengaruh Pengangguran Dan Inflasi Terhadap Jumlah Penduduk Miskin Di Provinsi Sumatera Utara Niagawan. 2022; 11: 168.
 Maipita and Fitrawaty M. Mengukur Kemiskinan & Distribusi Pendapatan Upp Stim Ykpn, Yogyakarta. 2014.
 Wijayanto and Dwi R. Analisis Pengaruh PDRB Pendidikan dan Pengangguran Terhadap Kemiskinan di Kabupaten Kota Jawa Tengah tahun. 2010; 20052008.
 Mudrajad K. Dasardasar Ekonometrika Pembangunan. UPP STIM YKPN. 2010.
 Suharto RB. Ekonomi sumber daya manusia. 2021.
 Susanto R and Pangesti I. Pengaruh Inflasi dan Pertumbuhan Ekonomi terhadap Kemiskinan di Indonesia. J App Bus Econ. 2010; 7: 271278.
 Imanto R, Panorama M and Sumantri R. Pengaruh Pengangguran Dan Kemiskinan Terhadap Pertumbuhan Ekonomi Di Provinsi Sumatera Selatan. AlInfaq: Jurnal Ekonomi Islam. 2010; 11: 118.
 Wu W, Wei HL, yang S and Xu Z. Nexus between financial inclusion, workers’ remittances, and unemployment rate in Asian economies. Hu Soci Sci Comm. 20123; 10: 110.
 Rahardja R and Manurung PDM. Pengantar Ekonomi (Mikroekonomi dan Makroekonomi). Jakarta: Lembaga Penerbit Fakultas Ekonomi Universitas Indonesia. 2008.
 Statistik BP. Statistik Sumatera Utara Sumatera Utara: BPS. Boediono. Seri Sinopsis Pengantar Ilmu Ekonomi No. 2 Ekonomi Makro Edisi Keempat. Yogyakarta: BPFE 2014; 2021.
 Santosa S and Agus B. Analisis Inflasi di Indonesia. Seminar Nasional Multi Disiplin Ilmu Unisbank 2017 Semarang Indonesia July 2017. Stikubank University. 2017.
 Sinaga JS. The Effect of BI Rate Exchange Rate Inflation and Third Party Fund (DPK) on Credit Distribution and Its Impact on NonPerforming Loan (NPL) on XYZ Commercial Segment Bank. Uni J Accou Finance. 2020; 8: 5564.
 Adhilla AN and Herianingrum S. Analisis Faktor Yang Mempengaruhi Kemiskinan Di Jawa Timur Perspektif Islam. Jurnal Ekonomi Syariah Teori Dan Terapan. 2020; 7: 1002.
 Tri AB and Nano P. Analisis Regresi Dalama Penelitian Ekonomi and Bisnis. 2016.
 Dkk ASR. Cara Cerdas Menguasai Eviews. Jakarta: Salemba Empat. 2011.
 Ariefianto A and Doddy M. Ekonometrika esensi dan aplikasi dengan menggunakan EViews Jakarta: ERLANGGA. 2012.
 Kependudukan DJ and Sipil DP. Catatan Sipil Sumatera Utara. Suamtera Utara: DUKCAPIL. 2021.
 Engle RE. And Granger CWJ. Cointegration and Error Correction: Representation estimation and testing Econometrica. 1987; 55: 251276.
 Gujarati and Damodar N. The McGrawHill Companies Basic Econometrics 4ed. 2004.
 Kasim R, Engka DSM, Siwu HD, Inflasi AP, Dan P, Pemerintah B and Siwu HD. Analisis Pengaruh Inflasi, Pengangguran Dan Belanja Pemerintah Terhadap Kemiskinan Di Kota Manado. Jurnal EMBA: Jurnal Riset Ekonomi, Manajemen Bisnis Dan Akuntansi. 2021; 9: 953963.
 Kolibuc M. 16456329761Sm. Jurnal Pembangunan Ekonomi dan Keuangan Daerah. 2017; 18: 114.
 Kuncoro M. Otonomi dan Pembangunan Daerah Reformasi, Perencanaan Strategi dan Peluang. Jakarta: Penerbit Erlangga. 2003.
 Lipsey L and Richard G. Pengantar Makro Ekonomi. Edisi Kedelapan, Jakarta: Erlangga. 2001.
 Mahsunah D. Analisis Pengaruh Jumlah Penduduk, Pendidikan Dan Pengangguran Terhadap Kemiskinan Di Jawa Timur. Jurnal Pendidikan Ekonomi (JUPE). 2013; 1: 117.
 Maipita M and Fitrawaty I. Mengukur Kemiskinan & Distribusi Pendapatan Upp Stim Ykpn Yogyakarta. 2014.
 Said R. Pengantar Ilmu Kependudukan Jakarta: LP3ES anggota Ikapi. 2012.
 Sembiring IPS, Simanjuntak S and Sitepu VA. Pengaruh Inflasi dan Pengangguran terhadap Penduduk Miskin di Sumatera Utara Tahun 20062020. Jurnal Ilmu Sosial Manajemen Akuntansi Dan Bisnis. 2021; 2: 113.
 Suripto S and Subayil L. Pengaruh Tingkat Pendidikan, Pengangguran, Pertumbuhan Ekonomi Dan Indeks Pembangunan Manusia Terhadap Kemiskinan Di D.I.Yogyakarta Periode 20102017. Growth Jurnal Ilmiah Ekonomi Pembangunan. 2020; 2.
 Susanto R and Pangesti I. Pengaruh Tingkat Pendidikan terhadap Kemiskinan di DKI Jakarta. JABE J App Bus Econ. 2019; 5: 340.
 Widarjono A. Ekonometrika Pengantar dan Aplikasinya, Edisi ke 4 Yogyakarta: UPP STIM YKPN. 2013.
 Widarjono A. Ekonometrika: Pengantar dan Aplikasinya. Edisi Ketiga. Cetakan Pertama. Penerbit Ekonesia Yogyakarta. 2009.
 Zurlinda Z. Analisis Pengaruh Pertumbuhan Ekonomi, Jumlah Pengangguran dan Indeks Pembangunan Manusia (IPM) terhadap Jumlah Penduduk Miskin di Kabupaten. Jurnal Ekonomi Hal. 2020; 9: 2936.