Demographics, Wealth, and Global Imbalances in the Twenty-First Century

Adrien Auclert, Hannes Malmberg, Frederic Martenet and Matthew Rognlie, "Demographics, Wealth, and Global Imbalances in the Twenty-First Century," National Bureau Of Economic Research, August 2021, https://www.nber.org/papers/w29161

Note Appendix F squares findings with existing results, “…we show that our results are useful to understand existing findings in the literature. First, across papers that conduct a similar exercise, we trace results back to their inputs, and show why different assumptions about the compositional effect are a critical driver of the differences in general equilibrium outcomes. Second, within papers that consider the role of parameter changes, we show that our results are useful in explaining the functional form relationship between these parameters and general equilibrium outcomes. In the interest of space, we focus on the effect of demographic change on the total return r (sometimes referred to as the natural interest rate, or r∗, in the literature)…”

"...One key observation in this literature is that the savings rate is hump-shaped in age, so that as the population continues to age, the aggregate savings rate eventually declines. Observers have made various macroeconomic predictions based on this effect: that aging will raise interest rates...decrease standards of living by impairing capital accumulation....or exert inflationary pressure as the number of consumers increases relative to the number of producers (Goodhart and Pradhan 2020). These predictions are not borne out in our analysis. Instead, we find that aging unambiguously lowers the real interest rate, thereby increasing capital intensity and output. A lower real interest rate also implies less inflationary pressure in any standard model in which this pressure is captured by the natural interest rate..... Figure 9 shows the resulting projected savings rates until 2100. These are indeed negative in all countries. In panel A of figure 10, which represents steady-state equilibrium between savings and investment, we depict this effect as a leftward shift in the private savings curve. At first glance, this might seem to imply an increase in r, as represented by the hollow circle. But since demographic change lowers the population growth rate and therefore g, the other curve—representing net investment and public borrowing—also shifts left, and the overall effect is a decline in r.To understand this result, it is useful to compare to panel B, which depicts steady state equilibrium between asset demand and supply. Here, only the asset demand curve shifts—to the right—and the unambiguous implication is a decline in r. But the curves in panel A are identical to panel B, just both multiplied by g. Hence, although both curves in panel A shift left, the net investment curve shifts left by more, producing the same decline in r as in panel B. We conclude that the “flow” view of equilibrium in panel A is in principle just as valid as the “stock” view of equilibrium in panel B, but only if we remember the effect of g on net investment. Ignoring this effect in the context of demographic change, which can significantly push down long-run g, may give the wrong sign for the change in r..."

".. .Table 1 presents our results. The left-hand panel shows the changes in the rate of return, calculated using equation (13), while the right-hand panel shows the average change in log wealth-to-GDP in percent, calculated using equation (14). We find that the equilibrium return r unambiguously falls in response to demographic change, refuting the “great demographic reversal” hypothesis...In our central scenario, r falls 123 basis points by the end of the twenty-first century; the fall is larger when σ or η are small, since this limits the responsiveness of asset supply and demand to falling returns. For our central scenario, average wealth-to-GDP increases by 10%, or approximately 47 percentage points in levels as a share of GDP..."

"...Next, we turn to the evolution of net foreign asset positions. Figure 6 shows the changes between 2016 and 2100 predicted by the formula in proposition 3. The bars display the main results, which feature a large divergence in NFA positions, with India and China experiencing increases of 44 to 125 percentage points and Germany experiencing a decrease of 55 percentage points.... Panel A of figure 7 implements this calculation. The solid lines show global imbalances until today for the five large economies discussed in the introduction, and the dashed lines show the projections from equation (25). In the next few decades, we expect to see a widening of existing global imbalances: China’s net foreign assets will rise substantially, while those of the US will decline. Although these trends flatten mid-century, the second half of the 21st century features a conspicuous rise in India’s net foreign assets, offset partly by a decline in Germany and Japan, whose demographic transitions at that point are nearly complete...."

"...The main mechanisms are summarized in figure 4. The grey bars show the evolution of the population distribution, starting young in 1950 and growing progressively older over time. In the figure, this population evolution is superimposed with the 2016 profiles of assets and labor income, with panel A illustrating how demographic change pushes up assets by moving individuals into high asset ages, and panel B illustrating how demographic change first pushes up aggregate labor income as the baby boomers reach middle age—the so-called “demographic dividend” (Bloom et al., 2003)—and later pushes down aggregate labor income as more individuals reach old age...Figure 5 displays the evolution of ∆ comp,a t and ∆ comp,h t (multiplied by W0/Y0 to obtain level effects on wealth-to-GDP).24 Panel A shows that ∆ comp,a t monotonically pushes up the wealth-to-GDP ratio throughout the sample period. The trend flattens towards the end of the 21st century as aging becomes concentrated in very old ages where asset accumulation ceases. However, the trend never reverses, due to the well-known fact that asset decumulation in old age is very limited....For other countries, the logic behind ∆ comp is broadly similar to that for the United States..."

The compositional effect of demographics

“…The results from this calculation are displayed in figure 2. Between 1950 to 2016, the compositional effect is positive in all countries, with an average increase of 80pp of GDP, and an increase of 105pp in the United States. These effects are quantitatively large. As a point of comparison, the actual changes in W/Y that occurred over this period were 220pp for the average country with available data in the WID, and 118pp for the US. Looking ahead from 2016 to 2100, the effects remain positive, are even larger on average, and are heterogeneous across countries, ranging from 48pp in Hungary to 237pp in China and 327pp in India, with a 147pp increase in the United States. In the high fertility scenario, the effect is reduced by a younger population: it is brought down to 75pp in China and to 142pp in the United States; in contrast, the low fertility scenario sees even sharper aging, and the effect swells to 245pp in the United States and 447pp in China. Figure 3 provides more detail on the heterogeneity across countries, with the solid bars displaying the predicted compositional change in W/Y to 2100 for the main population scenario. In principle, this cross-country heterogeneity could reflect either differences in demographic evolution or differences in the age profiles of assets and labor income. While both matter, the former is the main factor: countries with large effects are those whose demographic transitions are later and faster. The transparent bars in figure 3 illustrate this by showing similar cross-country heterogeneity in compositional effects if we counterfactually assume that all countries have the same asset and income profile as the United States...."

they note the impact of government debt can change this underlying dynamic, “…Although our baseline exercise holds government debt-to-GDP policy fixed, we show how rising government debt can mitigate or even undo the effect of demographic change on real interest rates, while increasing the effect on wealth-to-GDP…Since we have no direct way to predict the effect of demographics on long-run government debt targets, propositions 2 and 3 both assume a benchmark where each country keeps long-run debt-to-GDP constant. Appendix B.4 discusses alternative settings where debt-to-GDP changes in response to demographics. Two special cases stand out: when each country increases its debt-to-GDP target by the amount of its compositional effect, and when each country increases debt-to-GDP by the average world compositional effect. In the first case, there is no change in interest rates or net foreign assets, and each country’s wealth increases by exactly its compositional effect. In the second case, the same conclusions hold for interest rates and wealth, but net foreign assets in each country increase by the difference between its compositional effect and the global average, leaving the global imbalances predicted by proposition 3 intact….”

new Rognlie argues that Goodhart and Pradhan are wrong, there won't be a "great demographic reversal." He projects out the compositional effect of aging on the wealth-to-GDP ratio of 25 countries for the remainder of this century, bottom line"...According to our model, this will lead to capital deepening everywhere, falling real interest rates, and rising net foreign asset positions in India and China financed by declining asset positions in the United States. Our approach, based on stocks rather than flows, shows why there will be no great demographic reversal...."
"....we refute the great demographic reversal and show that, instead,demographics will continue to push strongly in the same direction, leading to falling rates of return and rising wealth-to-GDP ratios. We find that the key force is the compositional effect of an aging population: the direct impact of the changing age distribution on wealth-to GDP, holding the age profiles of assets and labor income fixed.In a baseline overlapping generations (OLG) model, this is a sufficient statistic for the actual change in wealth-to GDP for a small open economy. Further, for a world economy, the compositional effect— when aggregated across countries, and combined with elasticities of asset supply and demand that we obtain with other sufficient statistic formulas—fully pins down the general equilibrium effect on wealth-to-GDP, asset returns, and global imbalance..."

"...We measure the compositional effect by combining population forecasts with house-hold survey data from 25 countries over the period 2016-2100. We find that it is positive and large everywhere, but also heterogeneous, ranging from an increase in wealth-to GDP of 48pp in Hungary to 327pp in India. Since the average effect is positive and large, our model shows that there will be no great demographic reversal:through the twenty first century, population aging will continue to push down global rates of return, with our central estimate being -123bp, and push up global wealth-to-GDP, with our central estimate being a 10% increase, or 47pp in levels.Since the effect is heterogeneous across countries, our model predicts that demographics will also generate large global imbalances. For instance, we find that India’s net foreign asset position will steadily grow until it reaches 100% of GDP in 2100, while the United States’s net foreign asset position will decline to absorb this demand for assets...."

".. The average share of the population above 50 years of age has increased from 15% to 25% since the 1950s, and it is expected to rise further to 40% by the end of the twenty-first century(Figure 1, Panel A). There is a widespread view that this aging process has been an important driver of three key macroeconomic trends to date. According to this view, an aging population saves more, helping to explain why wealth-to-GDP ratios have risen and average rates of return have fallen (Figure 1, Panels B and C).1 Insofar as this mechanism is heterogeneous across countries, it can further explain the rise of global imbalances..."