Demographic supply-demand imbalance in industrial structure in the super-aged nation Japan

Abstract

Background: Japan has a rapidly decreasing population, with ultra-low fertility and extremely fast aging. The rapid dynamics constitute a warning that change in the industrial structure may be unable to meet the changing pace of age-dependent demand.

Methods: The present study estimated the supply-demand imbalance by industrial sector, and we investigated the effectiveness of possible countermeasures. To quantify the demographic burden of different industry experts, we employed the dependency ratio to calculate the supply and demand of each industrial sector and occupation.

Results: We identified an expected excess of demand in the health-care sector; the growth in that deficiency is likely to continue until 2045, when the elderly population is likely to reach a peak. By contrast, oversupply is expected in the education and construction sectors. An overall shortage of full-time workers is likely to continue until 2050, when we predict that Japan will lack 3.1-9.3 million full-time workers to satisfy the baseline demand level.

Conclusions: Considering that the imbalance is evident over different sectors, interministerial regulation of occupational choice may need to be imposed, e.g., by drastically changing student sizes in different area of higher education. Japan may have to decide to downgrade its social services and potentially consider increasing immigrant workers.

Keywords: Ageing; Dependency ratio; Mathematical model; Occupation; Population.

Fig. 1

Age composition and dependency ratios in Japan for 1920–2115. a Age composition of the population according to three different groups. b Youth, old-age, and total dependency ratios calculated as the ratio of the number of, respectively, children (up to 14 years), elderly (65 years or over), and both of those dependent populations to the working-age population (15–64 years). The number of those dependents per 100 working adults is indicated. The vertical line shows 2017, when our analysis was conducted. The population data by 2015 is based on population estimates (13); the data thereafter are population projections (14). The age-specific population size was unavailable for 1941–43 because of World War II. For the same reason, Okinawa was excluded for 1945–46

Fig. 2

Estimated dependent persons per industry expert in Japan. a Education sector from 1950 to the present day. Statistical datasets for kindergartens plus nursery schools were available only from 2002. K., kindergarten; Univ., university; E.S., elementary school; H.S., high school; K. + N.S., kindergarten and nursery; J.S., junior high school. b Medical and health-care sector. Dr., physician; Ns., nurse; Ph., pharmacist; E.T., emergency medical technician. Physician and pharmacist data were available from 1955; nurse and helper data are from 1960 and 1990, respectively. Japan started licensing emergency medical technicians in 2000. Statistical data were recorded every 5 years from 1955 to 1970 but every 2 years since 1970. c Social security sector. The data for the Self-Defense Forces are from 1955, firefighters from 1956, and police officers from 1981. d Infrastructure sector from 1953. 1) Agriculture and forestry; 2) transport and communication; 3) construction; 4) wholesale and retail trade, eating and drinking places; 5) manufacturing; 6) fisheries (right); and 7) electricity, gas, heat supply, and water (right). Data for electricity, gas, heat supply, and water were available from 1968. The industrial categories were revised in 2002—especially for transport and communication and wholesale and retail trade, eating and drinking places. A category bearing the legend “right” indicates that the vertical axis is measured on the right vertical axis owing to a different scale of dependence compared with other closely related populations

Fig. 3

Predicted labor force and labor state in Japan. a Linear and nonlinear predictions of labor force participation using the working-age population size from 1995 (upper right for both males and females) and up to 2015 (lower left for both males and females). Best fit models along with expected value equations with estimated parameters are shown inside the panel. b Estimated number of workers based on linear and nonlinear extrapolations of labor force participation from 2015 to 2065. c, d Age-dependent snapshot of the labor state for males and females, respectively, according to the 2015 census

Fig. 4

Predicted supply and demand of the labor force by industrial sector in Japan. The predicted supply and demand were computed by classification according to industrial sector using the average standard level for 2002–06. a Expected demand for industry-specific labor force to maintain the standard level. b Predicted supply of industry-specific labor force based on predicted worker size. Distribution of the industry-specific expert size was assumed to be maintained from that in 2015. c Predicted lack of experts, i.e., demand minus supply, and d relative proportion of redundant experts compared with the average for 2002–06

Fig. 5

Predicted excess of overall demand and scenarios for recruiting additional full-time workers. In both panels, the thick solid line shows the estimated excess of demand in Japan; it is predicted to reach a maximum of over 9 million people by around 2050. That excess is compared with possible compensation scenarios. a Forecast of labor force replenishment by change in rate of full-time homemakers. The percentage represents the proportion of homemakers that could theoretically be recruited for full-time employment. The baseline (status quo) is 0 people. Even with 100% recruitment, the additional force cannot meet growing demand by 2035. b Forecast of labor force replenishment by 10-year delay in retirement age. The baseline (status quo) is that everyone is advised to retire at age 60 years and 0 people. The scenario is when that age is extended to 70 years. Demand exceeds the additional labor force by 2025