Title:
FIND OUT THE RELATIONSHIP BETWEEN EDUCATION BUDGET AND UNEMPLOYMENT RATE IN TAIWAN BY THE AUTOREGRESSIVE DISTRIBUTED LAG MODELS

Authors:
Ko-Ming Ni, Taiwan

Abstract:

The relationship between the unemployment rate (%) and the education budget is discussed in this paper. The time-series data are subtracted from archives of the Directorate-General of Budget, Accounting and Statistics (DGBAS), Ministry of Education (MOE), and Ministry of Labor (MOL) in Taiwan. Totally 39 years (from 1981 to 2019) of data are scrutinized for analysis. Both the unemployment rate and education budget are not stationary checking by the augment Dickey-Fuller (ADF) unit-root test. The spurious result will be obtained if the ordinary least square method is used to perform the regression. The first difference of both time series are stationary and can be denoted as I(1). A further check the residuals of the regression by the cointegration test shows the residuals are not stationary (not cointegrated), then the more complicated autoregressive distributed lag (ARDL) models are chosen to solve this difficult problem. After analyzing two nonstationary and not cointegrated time series, the author finds that increasing the education budget can reduce the unemployment rate. The calculated unemployment rates are checked with the real ones, the average error of the past 39 years is amazingly less than 10.0%, more precisely, only 9.76%. Under the assumption of constant increment of the annual education budget, to increase the one-unit education budget (10¹¹ NT) will roughly reduce 0.313% of the unemployment rate in Taiwan.