Usefulness of the article: This article aims to shed light on the impact of demographic aging on inflation, as well as to summarize research and studies conducted on the subject. It shows by extension that the challenges associated with aging will have to be addressed in a world of low inflation and growth.
Summary:
- The theoretical framework used by proponents of inflationary demographic aging suffers from unconvincing assumptions.
- Despite the potentially contradictory effects of demographic aging on inflation, the empirical evidence in favor of the disinflationary thesis seems to largely prevail.
- These disinflationary forces, caused by the increase in the proportion of retirees in the population, would primarily result from a decline in aggregate demand and would be associated with declining growth rates, lower stock market valuations, and a deterioration in the transmission of monetary policy.
- Economic policies, which will be responsible for managing the issues related to demographic change in industrialized countries, will likely have to be conducted in a context of low inflation and deteriorating macroeconomic performance.
The profound demographic changes underway in developed countries will significantly alter the productive structure of these countries in the near future. While there is broad consensus that the relative increase in the number of retirees compared to the working population will lead to lower growth rates, the question of the impact on inflation is much more debated.
After presenting in the first part the various inflationary and disinflationary theories on the effect of demographic aging on inflation, this second part has a twofold objective: to examine the multiple theoretical elements and offer a critique where weaknesses can be identified, and to compare theory with empirical estimates by summarizing recent measurements on the subject.
While inflation will be determined in the future by multiple structural channels such as the distribution of value added between labor and capital, the level of inequality, imbalances between savings and investment, globalization, and wage dynamics, the demographic structure of each country will likely play a major role, as aging will alter many macroeconomic variables.
1. The inflationary thesis is based on unconvincing assumptions
While the arguments in favor of inflationary demographic aging are appealing, the assumptions underlying them are questionable.
First, the idea that households will deplete their savings at the end of their lives, in line with the life cycle, is not borne out by observation. Despite differences between countries, particularly in terms of the variety of pension systems, savings rates do not decline on average at the end of life. In France, the savings rate even tends to increase again after the age of 70, to levels equal to or higher than those of working households.
Figure 2: Savings rates of French households by age
Source: Bellamy V., Fesseau M., Raynaud E., « Inequalities between households in the national accounts: greater disparities in income than in consumption, » Insee Première No. 1265, November 2009.
This can be explained by the fact that households, regardless of their level of wealth, behave in a « dynastic » manner (motivation to pass on wealth to future generations) as they age, with very limited capital disaccumulation. The life cycle hypothesis is based on the work of Kotlikoff and Summers (1981), who show that intergenerational transfers are the main determinant of savings among older people. Retired households therefore do not contribute to stimulating demand but, on the contrary, inhibit it because their consumption is lower than that of other cohorts of the population on average. This phenomenon is reinforced by wealth disparities among retirees, which are significantly « more pronounced among retirees than among the working population » (Girardot-Buffard 2009). The share of the wealthiest retirees further fuels the increase in savings. Finally, the nature of retirees’ savings differs from that of working people: older people, who have a shorter life expectancy than people of working age, have a clear preference for liquid and short-term savings. In France, for example, the share of liquid savings among working people is around 40%, while this proportion exceeds 50% among retirees (Girardot-Buffard 2009). This shift in the structure of savings towards safer, lower-yield investments with age would penalize growth potential and lead to lower expected inflation.
Similarly, the idea that labor shortages, linked to the relative decline in the working-age population as a proportion of the total population, would facilitate the acceptance of wage demands does not take into account other institutional determinants in wage negotiations. In particular, it ignores the secular decline in union membership among employees in most developed countries since the late 1950s.
Table 1: Union membership rates among employees in developed countries in 1999, 2006, and 2018
Source: OECD, Negotiating Our Way Up report « Collective Bargaining in a Changing World of Work, »
calculations by BSI Economics
Ignoring changes in union membership rates can lead to an overestimation of workers’ ability to obtain wage increases. Indeed, the relationship between unionization rates and wage growth can be observed both within countries and in international comparisons. Without major changes in unionization levels, it is unlikely that wage inflation will take hold despite growing labor shortages.
Figure 3: Relationship between unionization rates and wage growth
Source: OECD, Negotiating Our Way Up report « Collective Bargaining in a Changing World of Work, »
calculations by BSI Economics
These two examples show how fragile the assumptions underlying the arguments of advocates of inflationary aging are and how they are not reflected in contemporary data.
2. Empirical studies tend to validate the deflationary effect of demographic aging
The famous « reflation » predicted by Juselius and Takats (2015, 2018) from 2010 onwards, which they consider to be the turning point in OECD countries, does not seem to be materializing. On the contrary, the annual rate of inflation has fallen by an average of 1.1 percentage points over the last decade (from an average of 3% per year over the period 2000-2008 to just 1.9% over the period 2010-2019).
Numerous empirical studies have sought to isolate the causality between demographic aging and inflation, controlling for other factors that influence price levels. Although most recognize the ambiguity of the theoretical effects of aging on prices, the majority conclude that there is a deflationary effect.
Regardless of the estimation models used [(Yoon et al. 2014, Gajewski, 2015)[1], autoregressive models (Bobeica et al. 2017, Aksoy et al. 2019)[2], or models involving macroeconomic feedback loops (Anderson et al. 2014)[3], the majority of empirical studies report that demographic aging is associated with deflationary pressures, the main channel for which is a decline in aggregate demand, which easily outweighs the inflationary pressures resulting from the reduction in the available labor force.
Thus, the relationship between the old-age dependency ratio and inflation appears to be very strong and holds true both between developed countries (Figure 4) and within the same country over a long period (particularly for countries where aging is already at an advanced stage, such as Japan and Italy, which are the two countries with the highest old-age dependency ratios in the world; Figures 5 and 6).
Figure 4: Relationship between dependency ratio and inflation
Source: OECD, BSI Economics calculations
Charts 5 and 6: Relationship between dependency ratio and inflation in Italy and Japan
Source: OECD, BSI Economics calculations
Conclusion and policy implications
After presenting the opposing theories on the impact of demographic aging on inflation in the first article, this second part offers a theoretical and empirical critique of the idea that aging is inflationary. The article focuses in particular on highlighting the lack of verifiability of inflationary hypotheses such as those made about the fall in the savings rate of seniors or the rise in wage demands of the working population, which are contradicted by observation.
The disinflationary forces generated by the decline in the working population would stem from several channels, the main one being the decline in aggregate demand, and would be associated with lower growth rates, lower stock market valuations, and a deterioration in the transmission of monetary policy.
These results are ultimately consistent with the idea that demographic aging would lower the natural interest rate and growth potential, making monetary policy an increasingly ineffective lever for bringing inflation back to its target. Only proactive fiscal policies can prevent central banks from being blocked in their actions by hitting the lower bound.
Population aging will be a major challenge for economic policymakers, as they will need to ensure the sustainability of public pension systems and healthcare financing in countries where the number of working-age people will be lower than the number of retirees, and where macroeconomic performance will be weaker.
Bibliography:
Aksoy, Yunus, Henrique S. Basso, Ron P. Smith, and Tobias Grasl. 2019. « Demographic Structure and Macroeconomic Trends. » American Economic Journal: Macroeconomics, 11: 193-222.
Anderson, D, D Botman and B Hunt (2014) « Is Japan’s Population Aging Deflationary? » IMF Working Paper 14/139, August
Bellamy V., Fesseau M., Raynaud E., « Inequalities between households in national accounts: greater disparities in income than in consumption, » Insee Première No. 1265, November 2009.
Bobeica, E., Lis, E., Nickel, C., and Sun, Y. (2017) “Demographics and inflation” ECB Working Paper Series 2006, European Central Bank.
Broniatowska, P. (2017). Population aging and inflation. Journal of Population Aging, forthcoming.
Gajewski, P. (2015) “Is aging deflationary? Some evidence from OECD countries” Applied Economics Letters, 22: 916–919
Girardot-Buffard, P. (2009) “The wealth of retired households,” INSEE, Household Income and Wealth (2009 edition)
Juselius, M and E Takáts (2015) “Can demography affect inflation and monetary policy?”, BIS Working Paper 485, February
Juselius, M and E Takáts (2018), “The Enduring Link between Demography and Inflation”, BIS Working Paper No. 722.
Kotlikoff L. and L. Summers (1981), « The Role of Intergenerational Transfers in Aggregate Capital Accumulation, » Journal of Political Economy, 89, 4, pp. 706-732
Yoon, J-W, J Kim and J Lee (2014) “Impact of Demographic Changes on Inflation and the Macroeconomy” IMF Working Paper 14/210 November
[1] Panel regression estimated including fixed effects and robust to inter-country heteroscedasticity and series autocorrelation.
[2] Autoregressive model (VAR) verifying the cointegration of the series (positive long-term relationship between inflation and the growth rate of the working-age population as a proportion of the total population in all the countries studied).
[3]IMF GIMF models with macroeconomic closure. This is a multi-country DSGE model to which the authors add variables related to demographic change.
