Time Series Modeling of Realized Cost and Revenue of National Health Insurance in Taiwan

• Main Authors: Ching - Fang Wu, 吳靜芳
• Other Authors: Yi - Chia Wang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/59023598514115541994

碩士 === 東海大學 === 經濟系 === 99 === This paper aims to explore the factors that influence the (nominal and real) cost and revenue of the Bureau of National Health Insurance (BNHI) in Taiwan from short to long run. As the BNHI provides data on both accrual basis and cash basis, almost all related studies tend to explore the time-series properties of accrual-based cost and revenue of the BNHI. However, this paper has sufficient arguments to believe that only the cash-based cost and revenue can reflect the realized deficit or surplus on which the BNHI and the Taiwanese government should focus. In addition, cash-based cost and revenue are the variables in the whole macroeconomic system, while accrual-based variables are mostly policy-oriented. Exploring influential macroeconomic variables on the cash-based cost and revenue of the BNHI is one of this paper’s major contributions in our ARIMA (Autoregressive Integrated Moving Average with Exogenous Explanatory Variables) modes. The cost and revenue of the BNHI are respectively measured by the medical benefits received by registered medical institutions and the premium paid by individuals, firms and the government. On the one hand, in addition to AR and MA terms, the BNHI’s premium revenue is significantly affected, from short to long run, by unemployment rate, the growth rate of population under age 15, the growth rates of population above age 65, marriage and divorce rates. On the other hand, the BNHI’s payments to registered medical institutions is affected, from short to long run, by the number of physicians per person, birth rate and labor force participation rate. Structural breaks such as raising individual premium rates from 4.25% to 4.55% after September 2002 also significantly influence both the BNHI’s cost and revenue. Based on our estimation results, the impulse response analyses at the end of this paper conclude that any one-off shock can almost persist in the BNHI’s premium revenue, while the BNHI’s payment to registered medical institutions shows decreasing influence from a one-off shock across time.