Can we predict where interest rates are headed?

Long-term real interest rates are low (Bean 2015). In many advanced economies the rate on long-term government bonds is negative once adjusted for expected inflation. In emerging markets, real rates are a little higher, but only because of heightened perceptions of sovereign risk since the ‘taper tantrum’ in 2013. This low global interest rate environment has not materialised overnight. Rather, it reflects a prolonged fall in long-term real interest rates over the past 30 years. Since the 1980s, market measures of 10-year risk-free real rates have declined by around 450 basis points (bps) across emerging and developed economies (Figure 1).

Figure 1. Comparison of real rates across the world

Sources: King and Low (2014); Consensus Economics; IMF; DataStream; authors’ calculations.

Although there is a lot of variation across countries, the presence of a discernible common trend in real rates suggests global factors are at work. Since the decline has largely occurred against a backdrop of low and stable inflation with little sign of global demand overheating, it suggests the sustained fall in long-term market rates is symptomatic of a fall in the global equilibrium rate of interest.

New research

Typically, this equilibrium rate is thought to depend on expectations of global trend growth and factors shaping preferences for desired savings and investment. In a new paper (Rachel and Smith 2015), we aim to analyse and quantify the effects of both. First we use a modified growth-accounting framework to interrogate the various secular trends that could be affecting global growth – including the ‘headwinds’ highlighted by Gordon (2014). Then, we use a simple saving-investment framework to analyse global shifts in desired savings and investment, illustrating how changes in preferences could have affected the neutral rate.

Growth

Although changes in global trend growth are probably the most commonly cited driver of changes in real interest rates, we find it difficult to account for much (if any) of the pre-Crisis fall in the global real rate by just appealing to past changes in growth; global growth was fairly steady in the pre-Crisis decades. However, the Global Crisis is likely to have triggered a reassessment of growth prospects going forward. Pessimistic expectations of future growth could now be playing an important role in driving the decline in real rates we have seen most recently. Our analysis suggests that slower global labour supply growth (due to demographic forces) and some headwinds at the technological frontier (such as a plateau in educational attainment), may cause global growth to slow by up to 1 percentage point over the next decade (see Figure 2).

Figure 2. Global growth accounting

Sources: TED, US Conference Board, IMF, UN and authors’ calculations.

This assessment happens to be broadly consistent with the revisions to long-term growth expectations seen since the Crisis (Figure 3). Using a calibration based on standard neoclassical growth theory, we think this decline could account for around 100bps of the fall in the global real rate we have seen recently.

Figure 3. Consensus forecasts for long-term growth

Sources: Consensus Economics and IMF WEO.

Preferences for desired saving and investment

Given growth expectations, the global equilibrium rate will also depend on agents’ preferences for desired savings and investment. Intuitively, desired savings will tend to rise as real rates increase because higher real rates generate higher returns and yield higher future consumption. By contrast, desired investment will tend to fall as real rates rise as it becomes more costly to invest. These two relationships describe the familiar investment-savings (IS) curves, which we use to frame our analysis. Neither curve is observable, of course – global saving must, by identity, equal global investment – we only observe the points at which the curves intersect. Interestingly, if we plot these intersections over time, a vertical pattern emerges; despite the 450 basis points fall in the global real rate over the past 30 years, global savings and investment have remained fairly stable (yellow diamonds, Figure 4).

Figure 4. The investment-saving framework

Sources: IMF, King & Low, 2014 and authors’ calculations.

This suggests that both curves have shifted. Various trends have been put forward to explain such shifts – some of them temporary (see, for example, Rogoff 2015) and some of them permanent (see, for example, Summers 2014). But few studies have attempted to bring together these different trends and quantify the relative importance of each in explaining the fall in the global real rate. We aim to do that here. We offer a taster of our analysis by outlining how demographic forces have affected global desired savings. We then briefly summarise the impact of other secular trends analysed more fully in Rachel and Smith (2015).

Demographic trends and global savings

Over the past 30 years the proportion of dependents (those aged 0-19 and 65+), has fallen from 50% of the global population to 42%. There is a stable negative relationship between dependency ratios and saving rates across countries over time (Figure 5).

Figure 5. Saving rates and dependency ratios

Sources: IMF WEO and UN Population Statistics.

This is consistent with the life-cycle hypothesis (Ando and Modigliani 1963) – those who work are also those who save. Every percentage point fall in the dependency ratio translates to around a 0.5 percentage points rise in national saving rates. So, the 8 percentage points fall in the global dependency ratio seen over the past 30 years should equate to a 4 percentage points increase in desired savings as a share of world GDP, i.e. a 4 percentage points right-shift in the saving schedule. Using estimates of the slopes of the saving and investment curves from the empirical literature, we find that this 4 points right shift in the saving schedule translates into an actual fall in the global real rate of around 90 basis points (Figure 6).

Demographic forces can thus explain around a fifth of the 450 basis points fall in the global real rate we have seen since the 1980s.

Figure 6. Shift of the desired saving schedule due to changes in the composition of the global population

Sources: IMF and King & Low, 2014 and authors’ calculations

Looking ahead, dependency ratios are expected to remain low this decade and then increase gradually as population ageing starts to bite. This should cause the effect of demographic forces on real interest rates to unwind over time, pushing up on real rates in the future. There is much uncertainty over the speed at which this unwind will occur. Some, such as Goodhart and Efurth (2015), suggest a quick turnaround. Our analysis points to a more delayed effect, suggesting the global real rate will remain low for some time.

In addition to the 90bps effect from demographics, our analysis suggests that higher inequality within countries and a glut of emerging-market saving have also pushed down on the global real rate (by 45bps and 25bps, respectively). If this had been the whole story, we would have expected to see a steady rise in observed saving rates globally. But as highlighted above, global saving and investment ratios have been remarkably stable over the past 30 years, suggesting desired investment levels must have also fallen. We pin this decline in desired investment on a fall in the relative price of capital goods (accounting for 50bps of the fall in real rates) and a preference shift away from public investment projects (20bps). We also note that the rate of return on capital has not fallen by as much as risk free rates – the rising spread between these two rates has further reduced desired investment and pushed down on risk free rates (by 70bps).

This savings-investment framework provides a broad description of the relative sizes of the different forces at play.

The confidence interval around such estimates is clearly very wide, but taken at face value, shifts in preferences appear to explain around 300bps of the decline in real rates since the 1980s (shown by the shifts in the saving and investment curves labelled 1-6 in Figure 7), on top of the 100bps explained by the deterioration in the outlook for trend growth (illustrated by a shift in the investment curve, labelledgin Figure 7).

In other words, we think we can account for the bulk of the 450bps decline in the global real rate by appealing to long-run secular trends.

Figure 7. Quantifying shifts in desired savings and investment

Moreover, using reasonable assumptions about the future direction of these underlying trends, we think that these effects are unlikely to turnaround quickly (Figure 8). This suggests that the global equilibrium real rate may settle at or slightly below 1% over the medium to long run.

Figure 8. Secular drivers of global real interest rates

Policy implications of persistently low interest rates

The policy implications of persistently low real interest rates are extensive. If the global equilibrium real rate is at or slightly below 1%, then for countries with a 2% inflation target, equilibrium nominal interest rates in individual countries may eventually settle at or below 3% – considerably lower than the historic norm. In the face of adverse shocks, central banks may therefore be more likely to run up against the zero lower bound on nominal interest rates, requiring the use of unconventional policy instruments such as quantitative easing (QE) more often. However, uncertainties over the transmission of QE and concerns over the size of central bank balance sheets might limit the use of such tools in the future. For large adverse shocks, fiscal policy may therefore need to bear more of the burden of business-cycle management. Low rates may also fuel search-for-yield behaviour, posing challenges for macro- and microprudential policymakers. Overall, the possibility of the global equilibrium rate remaining at persistently low levels should motivate a real debate across the policy spectrum on the best approach to stabilise the cycle in the future. We hope the analysis presented here helps fuel that debate.

References

Ando A and G Modigliani (1963) “The Life Cycle Hypothesis of Saving: Aggregate Implications and Tests”, American Economic Review Vol. 53, No. 1, Part 1 (Mar., 1963), pp. 55-84.

Bean C (2015), “Causes and consequences of persistently low interest rates”, VoxEU.org, 23 October.

Goodhart C A E and Erfurth P (2014) “Demography and economics: look past the past”, VoxEU.org, 4 November.

Gordon R J (2014) “The demise of US economic growth: restatement, rebuttal and reflections”, NBER Working Paper No. 19895.

Rachel L and T D Smith (2015), “Secular drivers of the global real interest rate”, Bank of England Staff Working Paper No. 571.

Rogoff K (2015), “Debt supercycle, not secular stagnation”, VoxEU.org, 22 April 2015

Summers L H (2014) “US economic prospects: secular stagnation, hysteresis and the zero lower bound”, Business Economics Vol. 49, No. 2