The authors apply new empirical techniques to determine how demographic transitions relate to economic growth and capital market returns. They use 60 years of data from 176 countries, and fit polynomials to the regression coefficients between age groups and both GDP growth and security returns. Instead of studying the nominal growth of aggregate GDP, they examine the more meaningful real per capita GDP growth adjusted for purchasing power parity. Similarly, rather than measuring annualized nominal returns, they exclude cross-country differences in national inflation rates and real risk-free rates by comparing excess stock and bond returns relative to each country’s domestic cash returns. This research design enables them to discern divergent patterns in the connections between demographic effects and economic growth, on the one hand, and financial markets, on the other.
The authors posit that total per capita GDP in an economy is the sum across age groups of the proportionate size of each age group in the general population times that group’s productivity per worker. This decomposition allows the authors to highlight the importance of material variation in productivity across age groups. Whereas older workers have the strongest impact on the level of GDP, young adults have a greater impact on the GDP growth rate than do middle-aged and older workers. Because they are starting from scratch, young workers’ productivity grows at a faster rate while they are acquiring experience. It is thus reasonable to expect per capita GDP growth to be stronger in emerging economies, where young adults are the dominant and ascendant group, than in developed economies where older workers and retirees are more prevalent.
The authors do not propose a new theory linking demographic changes with financial market returns. They do, however, refer to several published models that reflect common wisdom. Young adults are concerned with paying down student loans, setting up households, and starting families; their disposable income is directed almost entirely to consumption rather than saving. Investing in stocks, and later in bonds, becomes more important as people mature, advance in their careers, earn more, and start to prepare for retirement. Retirees, once out of the workplace, begin dissaving, liquidating assets (again, first stocks, then bonds), and spending pension benefits to support themselves.
Malthus’s study, An Essay on the Principle of Population, first appeared in 1798, and since that time researchers have published an extensive body of empirical work on the economic significance of demographic changes. The authors’ study is set apart, first, by their adopting the econometric methodology of fitting polynomial regressions that Ray C. Fair and Kathryn M. Dominguez introduced in the American Economic Review in 1991, and, second, by steps they take to increase the power of their statistical tests. These improvements include using five-year rather than annual rates of GDP growth and capital market returns, controlling for starting economic and valuation levels, using information from all age groups (15 five-year cohorts from 0–4 to 70+), and examining a large cross-section of countries. Extracting data primarily from the United Nations, Penn World Table, and Global Financial Data, the authors use more than 200 non-overlapping observations in their main tests and over 1,600 in their robustness tests.
Summarized by Philip Lawton, CFA