Determinants of U.S. health expenditure: Evidence from autoregressive distributed lag (ARDL) approach to cointegration

Vasudeva N. R. Murthy, Albert A. Okunade

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Using 1960–2012 annual time-series data for modelling, we apply the Autoregressive Distributed Lag Cointegration (ARDL) approach, to identify some major drivers of per capita real U.S. health spending. The ARDL Bounds testing procedure (Pesaran et al., 1999; 2001) has several econometric advantages compared with the standard Johansen and Juselius cointegration method. One distinguishing feature of this ARDL Bounds testing procedure is its ability to estimate the long-run economic relationship without requiring pre-testing the time-series for the presence of unit roots in the data generating process incorporated in the cointegration model. The empirical findings in this study indicate that per capita real income (INCOME), the population percent above 65 years (AGE) and the level of health care technology (HRD), measured as the level of Research & Development expenditure in health care are cointegrated. INCOME, AGE and HRD exert positive effects on U.S. health expenditure per capita. Unlike prior studies, this paper presents new empirical evidence indicating that the U.S. health care is a necessity, with an income elasticity estimate of around 0.92. We also find that medical technology advances play a major role in the long run rise of the U.S. health expenditure. We discuss implications of these findings.

Original languageEnglish
Pages (from-to)67-73
Number of pages7
JournalEconomic Modelling
Volume59
DOIs
StatePublished - Dec 1 2016

Fingerprint

Cointegration
Distributed lag
Health expenditures
Healthcare
Bounds testing
Empirical evidence
Econometrics
Expenditure
Testing
Income elasticity
Economics
Health spending
Unit root
Data generating process
Modeling
Medical technology
Time series data
Real income

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics

Cite this

Determinants of U.S. health expenditure : Evidence from autoregressive distributed lag (ARDL) approach to cointegration. / Murthy, Vasudeva N. R.; Okunade, Albert A.

In: Economic Modelling, Vol. 59, 01.12.2016, p. 67-73.

Research output: Contribution to journalArticle

@article{d7cb7f6593f34eea82c37ce3c452fee7,
title = "Determinants of U.S. health expenditure: Evidence from autoregressive distributed lag (ARDL) approach to cointegration",
abstract = "Using 1960–2012 annual time-series data for modelling, we apply the Autoregressive Distributed Lag Cointegration (ARDL) approach, to identify some major drivers of per capita real U.S. health spending. The ARDL Bounds testing procedure (Pesaran et al., 1999; 2001) has several econometric advantages compared with the standard Johansen and Juselius cointegration method. One distinguishing feature of this ARDL Bounds testing procedure is its ability to estimate the long-run economic relationship without requiring pre-testing the time-series for the presence of unit roots in the data generating process incorporated in the cointegration model. The empirical findings in this study indicate that per capita real income (INCOME), the population percent above 65 years (AGE) and the level of health care technology (HRD), measured as the level of Research & Development expenditure in health care are cointegrated. INCOME, AGE and HRD exert positive effects on U.S. health expenditure per capita. Unlike prior studies, this paper presents new empirical evidence indicating that the U.S. health care is a necessity, with an income elasticity estimate of around 0.92. We also find that medical technology advances play a major role in the long run rise of the U.S. health expenditure. We discuss implications of these findings.",
author = "Murthy, {Vasudeva N. R.} and Okunade, {Albert A.}",
year = "2016",
month = "12",
day = "1",
doi = "10.1016/j.econmod.2016.07.001",
language = "English",
volume = "59",
pages = "67--73",
journal = "Economic Modelling",
issn = "0264-9993",
publisher = "Elsevier",

}

TY - JOUR

T1 - Determinants of U.S. health expenditure

T2 - Evidence from autoregressive distributed lag (ARDL) approach to cointegration

AU - Murthy, Vasudeva N. R.

AU - Okunade, Albert A.

PY - 2016/12/1

Y1 - 2016/12/1

N2 - Using 1960–2012 annual time-series data for modelling, we apply the Autoregressive Distributed Lag Cointegration (ARDL) approach, to identify some major drivers of per capita real U.S. health spending. The ARDL Bounds testing procedure (Pesaran et al., 1999; 2001) has several econometric advantages compared with the standard Johansen and Juselius cointegration method. One distinguishing feature of this ARDL Bounds testing procedure is its ability to estimate the long-run economic relationship without requiring pre-testing the time-series for the presence of unit roots in the data generating process incorporated in the cointegration model. The empirical findings in this study indicate that per capita real income (INCOME), the population percent above 65 years (AGE) and the level of health care technology (HRD), measured as the level of Research & Development expenditure in health care are cointegrated. INCOME, AGE and HRD exert positive effects on U.S. health expenditure per capita. Unlike prior studies, this paper presents new empirical evidence indicating that the U.S. health care is a necessity, with an income elasticity estimate of around 0.92. We also find that medical technology advances play a major role in the long run rise of the U.S. health expenditure. We discuss implications of these findings.

AB - Using 1960–2012 annual time-series data for modelling, we apply the Autoregressive Distributed Lag Cointegration (ARDL) approach, to identify some major drivers of per capita real U.S. health spending. The ARDL Bounds testing procedure (Pesaran et al., 1999; 2001) has several econometric advantages compared with the standard Johansen and Juselius cointegration method. One distinguishing feature of this ARDL Bounds testing procedure is its ability to estimate the long-run economic relationship without requiring pre-testing the time-series for the presence of unit roots in the data generating process incorporated in the cointegration model. The empirical findings in this study indicate that per capita real income (INCOME), the population percent above 65 years (AGE) and the level of health care technology (HRD), measured as the level of Research & Development expenditure in health care are cointegrated. INCOME, AGE and HRD exert positive effects on U.S. health expenditure per capita. Unlike prior studies, this paper presents new empirical evidence indicating that the U.S. health care is a necessity, with an income elasticity estimate of around 0.92. We also find that medical technology advances play a major role in the long run rise of the U.S. health expenditure. We discuss implications of these findings.

UR - http://www.scopus.com/inward/record.url?scp=84978805733&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978805733&partnerID=8YFLogxK

U2 - 10.1016/j.econmod.2016.07.001

DO - 10.1016/j.econmod.2016.07.001

M3 - Article

AN - SCOPUS:84978805733

VL - 59

SP - 67

EP - 73

JO - Economic Modelling

JF - Economic Modelling

SN - 0264-9993

ER -