Report Contents

November 17, 2020

Macro Series

Comment: The importance of inflation expectations

BY Russell Jones

Share on twitter
Share on whatsapp
Share on facebook
Share on linkedin
Share on email
Share on reddit

Listen to our reports with a personalized podcasts through your Amazon Alexa or Apple devices audio translated into several languages

( 9 mins)
  • Inflation has been under control for several decades and is historically low.
  • But opinion is now divided: some fear enduring ‘lowflation’; others price instability.
  • Much will depend on the future dynamics of inflation expectations.
  • For now, inflation expectations are muted; indeed perhaps too muted for comfort.
  • This points to continued low, indeed negative, policy rates and further fiscal largesse.
  • But this may become excessive, sowing the seeds of a very different environment.

Inflation is low, but opinion on the outlook is divided

Divergent views

Inflation is currently historically low across both the advanced, and the emerging and less developed, economies, frequently significantly undershooting mandated central bank targets.

That said, there is much debate about the outlook. Opinion amongst the cognoscenti is split, and perhaps more than is usually the case. Some see few if any immediate risks of a major acceleration in inflation, given an environment of copious excess capacity, cowed workers, and heightened uncertainty. Indeed, the bigger threat may be a further extended period of uncomfortably low inflation, if not deflation. Others, however, fret about rapidly rising budget deficits and public sector debt burdens, increasingly unconventional central bank policy, a recent surge in monetary growth, excessive monopoly power, and the threat of enduring damage to productive potential.

Price dynamics

Analysis of inflation is prone to over-simplification …

No doubt, the economics of inflation can be complex, and in attempting to understand it there is a temptation, as with any complex process, to simplify. And with simplification comes the risk of oversimplifying. The reality, however, is that inflation is not just a matter of a single, simple equation with the rate of change in some measure of the money supply on one side, and the rate of change in prices on the other. To understand the inflation process, it is necessary to consider how prices are determined, including the factors that determine the pricing power of both firms and individuals. These are influenced importantly by the institutional architecture and structure of an economy, both of which constantly evolve; by recent experience; and by perceptions of what the future holds.

Within this overall framework, it can be argued that there are four over-lapping and interconnected areas of consideration that dominate the inflation process, each of which can, and will, exert a varying influence over time. These are: supply shocks; economic slack; inflation expectations; and inflation persistence. Perhaps the most important of these is inflation expectations, especially over the longer term.

The future and the here and now

… but expectations play a big role in its determination

Rising inflation expectations can be expected to intensify the pressure for compensatory wage and other cost increases, cause firms to seek to build a certain amount of ‘fat’ into their profit margins, and thereby raise actual inflation; and conversely.

Inflation expectations become more benign the more enduringly they remain stable and embedded in the decision-making processes of individuals, trade unions, firms, and governments. For this reason, central bankers are always keen that inflation expectations remain well-anchored around specified inflation targets. This makes the job of macro-stabilisation policy easier: in particular, it affords policymakers greater leeway to manipulate real interest rates, and thereby final demand, in response to ebbs and flows in the pace of economic activity.

Through a glass darkly

That said, inflation expectations are difficult to measure or model accurately. In general, however, they appear to retain an adaptive quality, being driven by peoples’ and firms’ experiences, and by what central bankers say, do and achieve. Low inflation credibility on the part of policymakers is of crucial importance in keeping inflation expectations and inflation low.

Inflation expectations can be measured in many ways

Inflation expectations can be estimated via market-based metrics, or via surveys of actual sentiment. Central banks tend to favour market-based approaches, because they reflect actual investment decisions or, in the vernacular, people ‘putting their money where their mouth is’.

The approach usually taken is to focus on ‘inflation breakevens’ – the difference between the yields on conventional sovereign bonds and inflation-adjusted securities of similar maturities.[1] The favourite variant for many central bankers is the five-year breakeven, five-years forward, the logic being that over five years the vast majority of cyclical and commodity price effects could be expected to have played out.

Central banks often favour market-determined variants …

Alternatively, inflation swaps can be used. These are agreements between two counterparties to swap fixed rate payments on a notional principal amount for floating rate payments linked to an inflation index, such as the consumer price index. These derivatives are used to transfer inflation risk from one counterparty to another.

A major problem with the breakeven approach is that inflation-adjusted securities contain rarely offsetting biases: that is to say, an inflation risk premium that compensates investors for the degree of inflation uncertainty, and a liquidity premium, that reflects the fact that there are relatively few of these securities in most jurisdictions, and those that are issued tend to appeal to ‘buy and hold’ investors.

… although these are not without their shortcomings

The use of ‘off the run’, or rarely traded, conventional securities can address at least part of the issue of liquidity premia, but it has been estimated that the inflation risk premium can result in an upward bias of up to 100bps in breakevens, with the inflation risk premium from forward breakeven variants more volatile.

The most common alternative method to measure inflation expectations is to rely on surveys of the population. For the US, for example, the University of Michigan publishes surveys of inflation expectations for the next year, and for the period five to ten years ahead. As it turns out, the former tend to be quite volatile, driven by the most recent short-term inflation trends. The latter tend to be more stable. However, policymakers have tended to be rather sceptical about surveys that are conducted over the telephone, and use relatively small samples, and these exercises do throw up anomalies, such as evidence that women tend to perceive inflation to be consistently higher than do men.

Surveys of opinion provide an alternative approach …

Also employed are surveys of economists and professional forecasters. In the US and UK at least, these have a long history, although clearly in formal inflation-targeting countries they will tend to involve considerable circularity. Also, inflation is difficult to model much beyond a two-year horizon, because beyond that period econometric models tend, by construction, to revert to trend.

… but these too have their problems

Finally, there are business surveys of prices paid, or prices charged, although these tend to be relatively short term in purview; and diffusion indexes rather than measures to gauge the magnitude of price changes. Nor do they typically explicitly factor in wholesalers’ or retailers’ margins.

Comment: The importance of inflation expectations 1

Perhaps a holistic approach is best

Overall, the wisest approach would appear to be to keep an eye on a range of measures, and to respond when an overwhelming majority are sending a consistent signal. In the US, for example, the Cleveland Fed reports estimates of the expected rate of inflation up to 30 years ahead, along with approximations of the inflation risk premium, the real risk premium, and the real interest rate, based on a model that embodies different types of Treasury security yields, actual inflation data, inflation swaps, and survey-based measures of inflation expectations.

The state of play

Although inflation expectations are often volatile from period to period, they have been soggy since the Global Financial Crisis, not returning consistently to the levels that prevailed before 2008. They had also been subsiding fairly consistently since the middle of 2018.

As the pandemic hit, they fell sharply, but then rebounded as initial lockdowns were lifted. The US five-year breakeven, five years forward, dropped quickly to a low of 0.85%, before returning to its pre-COVID range of 1.70-1.90%.

At the current juncture inflation expectations are very muted …

The Cleveland Fed’s composite measure of 10-year inflation expectations also dipped as the virus struck, falling to a nadir of 1.16%, before recovering to 1.34% in October. Its rebound has so far been rather less impressive than that of breakevens. Moreover, the inflation risk premium has shown a tendency to drop towards zero, which prior to 2019, it had never done before.

At least to date, there is little evidence that the inflation expectations anchor is dragging in such a way as to indicate a major move away from the presumption of low inflation of recent decades. Indeed, the biggest concern appears to be that inflation expectations are evolving in such a way that inflation will exhibit a tendency to continue to undershoot central bank mandated targets.

… and there appears to be little threat to price stability

Of degrees of policy freedom and too much of a good thing

The implication therefore is that central bank policy rates and interest rate term structures more generally will remain historically low in the short term. If so, this will provide continuing support for risk assets, including investments in real estate. Indeed, the danger may be of intermittent bouts of excess, or ‘bubbles’ in these markets.

At the same time, persistently low inflation expectations, and interest rates will afford governments greater latitude to pursue expansionary fiscal policies, especially as they seek to sustain incomes and jobs until the pandemic is past. But it is further to be expected that, with private investment a primary casualty of the COVID-19 recession, low sovereign borrowing costs will also encourage governments to launch new public investment programmes.

This points to a further period of macro policy largesse

The question is whether at some stage all this monetary and fiscal largesse becomes excessive, encouraging not just asset price inflation, but excess aggregate demand and a marked acceleration in the rate at which goods and services prices increase. This threat may for now seem to be rather distant, but it is far from inconceivable.

Inflation expectations should be closely monitored in the months and years ahead. They will offer an important early warning of any fundamental change in the inflation environment.

Watch fors

  • Continued historically-low inflation over the next few years.
  • Further cuts in central bank policy rates into negative territory and unconventional initiatives.
  • Continued fiscal activism to sustain aggregate demand.
  • Excessive macro stimulus ultimately encouraging inflation expectations anchors to drag higher.
  • The development of a wage-price spiral over the longer term.
  1. They are called breakevens because, were inflation at the rate implied by the disparity in the two yields, the realised rates of return for both types of security would be identical.

More by Russell Jones