Investment conclusions: Inflation expectations and scenarios for the next 20 years
As we’ve seen, some of the structural forces that shape inflation are shifting in interesting ways. But the balance of evidence does not suggest a significant likelihood of either a profoundly inflationary or profoundly deflationary scenario. We’re sorry if this makes the report a bit of a let-down.
In other words, inflation expectations are likely to remain well anchored and under control. Although monetary policy is abused constantly in the media, it is to its credit that long-term inflation expectations have remained so well anchored throughout the roller-coaster ride of the past 10 years.
Table 1 shows a range of scenarios that might play out over the next 20 years or so. It is possible to imagine innumerable nuances to the scenarios set out here, but we believe too many scenarios confuse more than they guide. Unsurprisingly, given what we have just said, we assign the highest probability to a benign scenario in which inflation expectations remain well anchored and any inflationary effects from ageing or deflationary effects from technological change are offset with monetary policy.
We assign the second-highest probability to the prospect of technologically induced low inflation. Ordinarily, technological change improves productivity, which ultimately drives wage growth over the long run (Krugman, 1994). Further, higher productivity lowers the production costs of goods and services, which stimulates demand. For example, if a TV now costs a third of what it did last year, most people will buy a TV and a new coat. Although there may be a lag, there is no lasting impact on inflation. Remember, as we set out in our introduction, inflation or deflation is a general rise or fall in prices — the net effect on all prices is key.5
Table 1 shows a few scenarios where we believe that might not happen — in other words, where new technologies increase supply but also lower aggregate demand. For more information on some of the technological change touched upon, we’d encourage you to read our recent report, How soon is now? The investment impact of disruptive technologies.
We assign the lowest probabilities to secular stagnation and ‘stagflation’ scenarios. We have set out why we think the secular stagnation thesis is unlikely to play out on page 11, but it is far from an impossibility. Stagflation — low economic growth and frequent recessions combined with very high inflation (such as the 1970s) — is also unlikely but the risk has increased under President Trump. Although Trump is unlikely to oversee all the policies set out in table 1, he may topple a protectionist domino that leads to populist/nationalist responses down the line. Protectionism and a global trade war are likely to cause higher inflation initially before segueing into something more akin to secular stagnation (read our report, Trade of the century: Popular politics and the hidden costs of protectionism, if you’d like a fuller explanation).
In the final column of table 1, we set out our thoughts on what the inflation scenario might mean for long-term investment strategies. As you can see from figures 10 and 11, equities (since 1980) only tend to struggle when inflation falls below 0.5% or rises above 3.5%. Unanticipated changes in inflation, however, as measured by Citi’s inflation surprise index, do seem to have an adverse effect on valuations (figure 14).
Financial assets are forward-looking. They are the present value of future cash flows. In figures 12 and 13 we show how they respond to changes in inflation expectations, but note that the response is highly conditioned on the accompanying change in growth expectations. We proxy growth expectations with the change in real yields on the 10-year government bond. The bars in figures 12 and 13 show the average monthly performance of a variety of assets as inflation and growth expectations rise or fall. Equities do very well when inflation and growth expectations both move higher — and the riskier the better. The worst scenario for all assets other than gold is when both inflation and growth expectations fall together. We must point out, however, that this analysis is based on the past 25 years (as far back as the data on inflation expectations allows), and as such reflects performance during rather benign inflationary conditions.
Recognising the opportunities
Lastly, as long-term investors, we are constantly on the lookout for investment themes that may provide a source of return irrespective of the vicissitudes of the economic cycle. We have touched on a number of such themes in this report — ageing, personalised medicine and automation, for example. If you would like more information on these themes, please contact your portfolio manager.
We’ll leave you with an interesting observation about how ageing might directly influence stock market valuations. Two professors of finance in the US hypothesised that if ageing affects the rate at which savings are accumulated and decumulated, ageing therefore directly affects the supply and demand of equity markets. If there are more people in saving mode accumulating financial assets than there are people dissaving and cashing in their portfolios, this will impact equity market valuations.
As such, the professors observed a clear correlation between the number of 40- to 49-year-olds relative to the number of 60- to 69-year-olds (or savers versus dissavers) and the price-to-earnings ratio of the US stock market — a simple measure of valuation (Liu & Siegel, 2011). Projecting the trend forward doesn’t make for happy reading (figure 15). We suspect lower investment returns, lower rates of saving and later retirement means the number of 40- to 54-year-olds versus the number of 65- to 74-year-olds is the more appropriate ratio today. And that makes a big difference. We also can’t establish nearly such a strong correlation outside of the US. Still, it’s an interesting theory, and certainly a reminder of how so many themes, all too often discussed in isolation, are inextricably connected.
5. There is evidence ecommerce and online price aggregators may lower inflation because they improve ‘price discovery’. One study has found that changes in the number of people looking online (not necessarily buying online) has lowered non-energy, non-food manufactured goods by 0.1% a year over the last ten years across Europe (Ciccarelli & Osbat, 2017). However the net effect on the general levels of prices is unclear.
Figure 10: Equity valuations in different inflation environments
Figure 11: Real returns and real earnings growth by inflation
Figure 12: Equity valuations in different inflation environments
Figure 13: Read returns and real earnings growth by inflation
Figure 14: Valuations and surprise inflation
Figure 15: Valuations and a saving versus dissaving demographic profile