Select your investor profile:

This content is only for the selected type of investor.

Individual investor?

Multi Asset

Inflation awakes

February 2018

Patrick Zweifel, Chief Economist

After years of lying dormant, inflation is coming back to life in the US, and at a pace that's not fully appreciated by investors.

The US is finally shaking off the last effects of the global financial crisis and the world economy is returning to health. Both of these point to accelerating US inflation in the near term, and by more than investors might think – though recent market volatility suggests they're waking up to the risks.

And as factors that held back inflation fall out of the price index calculations – not least commodity prices and exchange rate moves – there is a significant risk that the personal consumption expenditure (PCE) index, the US Federal Reserve’s favoured measure of price pressures, will climb sharply over the coming quarters.

Based on an analysis of 30 factors in five categories that we consider to be primary sources of inflationary pressure, we estimate a more than one in four chance that inflation will overshoot the Fed’s 2 per cent target this year, the highest probability since 2014.

inflating risks

Probability of a core inflation rate rising above 2%

inflation phases probability threshold

*Core PCE inflation >2% and rising (ie 1-year change in inflation is greater than 0) **R2 weighted average of probabilities of inflation estimated independently using 30 factors ***Determined as the average of estimated inflation probabilities since 1986.
Source: Pictet Asset Management, CEIC, Datastream, Bloomberg; data from 01.01.1970 to 01.02.2018.

The Phillips curveball

To see why inflation is coming back, it’s worth considering why it disappeared in the first place. The answer can be found in the unusual economic circumstances that prevailed at the start of the global financial crisis.

Typically, once unemployment drops below a certain level or the so-called output gap – the difference between an economy’s actual and potential output – is closed, inflation starts to rise. In part, that’s because employers compete for increasingly scarce workers by raising wages, which, in turn, are spent on goods. After a certain point, more spending leads to rising prices. This is what’s known as the Phillips curve relationship. Because this relationship hasn’t held lately, some economists speculate that it is becoming irrelevant – the US PCE index excluding food and energy rose just 1.5 per cent on the year in December, which is well below its 2 per cent target, even though unemployment is near the bottom of its historic range.

We would argue that there are two clear explanations Phillips curve’s strange recent behaviour. The first is very low levels of inflation at the start of the last recession and the second is globalisation. 

when inflation's low, us phillips curve only kicks in when the economy's running hot

US Phillips curves 1960-2017 when inflation rate  is below 2.5%

us inflation rate output gap
*Average of 2 approaches: 1) HP filter; 2) Recursive HP filter. The output gap is lagged by 5 quarters. Source: Pictet Asset Management, CEIC, Datastream. Data from 01.01.1960 to 01.01.2018.

When the global financial crisis struck in 2008, inflation was running at historically low levels compared to the start of previous economic shocks.

This made it more difficult for businesses and employees to react to weakening demand in the customary way – delaying price rises and reining in wage claims. Normally, recessions end when goods and workers become relatively cheap enough to entice demand. The higher the initial rate of inflation, the less time it takes for this adjustment to take place.

Low starting inflation makes the adjustment much slower. People could speed it along by cutting prices but this is psychologically painful – in other words, prices tend to be stickier on their way down than on their way up, which is one reason why economies rarely slip into outright deflation.1

As a consequence of these sticky prices, the Phillips curve started to do something odd.2

For periods when inflation runs at above 2.5 per cent, the Phillips curve has behaved the way theory said it should – a stronger economy is associated with faster price rises and vice versa. But when inflation was below 2.5 per cent and unemployment was high (which is to say there was a negative output gap), the Phillips curve flattened – prices failed to fall even though labour was abundant and growth was weak.

A global phenomenon

But low starting inflation is just one factor explaining the Phillips curve conundrum. Globalisation is another.

When William Phillips first described the relationship between unemployment and wages in 1958, he looked at an individual economy – the UK – in isolation. And that’s generally how the Phillips curve has been analysed since. But at a time when goods and labour markets are becoming more global, a different approach is needed. In a world of just in time delivery that crosses not just borders but continents, it’s no longer sensible to think purely in terms of local business conditions. Global supply chains and ever more trade between countries means that potential output has to be considered against an international perspective.3 So one country’s labour shortfall can be filled by another’s.

Although the trend towards globalisation has stuttered during the past decade, there is still a strong case to be made for factoring in international effects when discussing the Phillips curve. We looked at two periods, 1970-93, when there was relatively less globalisation, and after 1994 by which time communism and autarky were abandoned across most of the world. For each of the periods we analysed the US Phillips curve against a US output gap and against a global output gap. We found that US wages responded to US unemployment when the country was less open to global trade. By contrast, they were relatively more sensitive to the state of the international economy after globalisation took off. 

international factors trump domestic ones in Us phillips curve 

US Phillips curve based on domestic and international output gaps over two separate periods

US core inflation rate
*Imports-weighted output gaps (41 US trade partners) **Average of 2 approaches: 1) HP filter; 2) Recursive HP filter. The output gap is lagged by 3 quarters for domestic and 1 quarter for international. Source: Pictet Asset Management, CEIC, Datastream. Data from 01.01.1970 to 01.02.2018.

The significance is that the US’s increasing internationalisation since 1994 means that US inflation is no longer just a domestic phenomenon – it is increasingly influenced by global factors.

So falling US unemployment has less follow-through to domestic price pressures. And by the same token, a shrinking global output gap will have relatively more of an impact on US inflation.

Coming at you from two directions

The US economy has worked its way through past deflationary pressures. After nearly a decade of low but positive inflation, the gap between where prices ought to be and where they are has largely evaporated.

What is more, the world’s output gap has also almost disappeared amid solid global growth, which is likely to exert more of an effect on US inflation. Among the early signs are likely to be relatively sharp recent increases in US wages and the sharp jump in commodity prices, led by oil.

Investors and policymakers have become so habituated to low inflation that this reversal could come as an unpleasant surprise. Inflation hasn’t died. It’s just been sleeping.