Abstract
Using quarterly data from 1994 through 2013, this article is the first to use the Hansen and Seo (2002) tests to examine the long-term relationship between real per-capita GDP (PGDP) and real per-capita health care expenditure (PHCE) in Japan with threshold effects. We detect the presence of a threshold cointegrating relationship between the two variables. In addition, we find that error correction adjustments are significantly conducted only through PGDP in a typical regime, whereas both PGDP and PHCE drive the adjustments in the extreme regime. We find that the extreme regime occurred mainly after Q4 2008, implying that the policy introducing the late-stage medical care system for the elderly in April 2008, which expanded the ratio of personal expenses for the elderly, may be effective for attaining a long-run equilibrium between PGDP and PHCE.
Notes
1 The ‘typical’ regime and the ‘extreme’ regime comprise approximately 83% and 17% of all observations, respectively.
2 We assumed that 80 grid points for the threshold and vector.
3 The ‘universal insurance system’, under which everyone can receive medical services at any medical institution nationwide in Japan, was developed in 1961.