A rule of thumb of mine (the O’Sullivan Zombie Rule of Economic Modelling) states that once a model has been written off as dead by the economics profession, it makes a comeback. The ‘death of budget deficits (under Clinton)’, the ‘death of value investing’ and a ‘this time is different approach’ to debt are some examples.
The latest one I have in mind is the Phillips Curve – an economic relationship researched by the New Zealand economist Bill Phillips that maps an inverse relationship between unemployment and inflation, and subsequently developed by prominent economists like Milton Friedman and Robert Lucas.
Over the past decade, a period characterised by low inflation, low interest rates and low unemployment, a number of economists have sketched the obituary of the Phillips Curve. James Bullard, a prominent Federal Reserve official, has stated ‘If you put it in a murder mystery framework – “Who Killed The Phillips Curve?”– it was the Fed that killed the Phillips curve’. Peter Hooper and Frederic Mishkin have pondered ‘The Phillips Curve – dead or Alive’, while a 2022 discussion paper from the Fed wondered ‘Who Killed the Phillips Curve? A Murder Mystery’.
There are some good reasons as to why the death of the Phillips Curve has been declared – falling rates of unionisation in the US have diminished the bargaining power of workers. In the UK in the 2010’s the sharp increase in the gig economy – where many workers effectively privatised themselves – also meant that a large number of workers had little wage bargaining power. An environment of generally falling productivity also laid bare the weaker claim that workers had to higher wages.
Though empirical evidence suggested that in many countries the Phillips Curve is dead, it still remains an important and well-worn policy setting for the major central banks, and many of them devote considerable resources to researching them as this paper from the ECB shows.
For instance, in recent years, in the US Janet Yellen as Fed Chair repeatedly spoke of driving down long-term unemployment to help spur a little inflation. Central bankers are typically very conservative, slow moving creatures – hence the logic of my ‘Zombie Rule’ is that by the time they reject a model, it is time to bring it back.
Fed behind Curve
The reason I think this to be the case now, is that many developed economies are perched between multi-decade highs in inflation, and multi-decade lows in unemployment. The prevailing view is that inflation is now decelerating, and strong employment means that we will experience a ‘soft landing’. This appears to be the view that financial markets are extrapolating from recent comments by the Federal Reserve Chair Jerome Powell.
In this context, the risk is that the Phillips Curve makes a Lazarus-like comeback in policy circles, and in practical terms that tight labour markets lead to very sticky, high inflation. One mystery in this respect is the ways in which the labour market is changing because of demographics, the post COVID economy and the attendant changes in the geographic location of labour, as well as the impact of ‘strategic competition’ on supply chains and thus labour markets.
Most of these factors however should lead to upward pressure on wages and it is striking that economies that have seen high inflation, are those where labour market participation has changed. To that end we will very likely hear more about the ‘revival’ of the Phillips Curve as we enter into a highly noisy macro environment characterised by highly pessimistic readings from lead indicators and very competitive labour markets.
A clue to how this plays out may come from my favourite piece of work from Bill Phillips.
Well before he was celebrated for the ‘Curve’, Phillips built an extraordinary machine that pumped colored water through glass vessels in order to demonstrate how money flows around an economic system. The machine named MONIAC (monetary national income analogue computer) included parts salvaged from a Lancaster bomber. Levers in the machine permitted users to simulate the effect on the system of fiscal (budget) policy changes for example, and such was the intuitive appeal of the machine that major universities like Harvard and Oxford ordered their own versions.
In today’s algorithmic driven economies and markets, such a simple contraption might seem well out of place, but the time might be ripe for the major central banks to install MONIAC’s.
At very least their use might help induce further humility to a central banking community that has gotten the inflation call badly wrong in the past two years, and that is now presiding over a premature easing in financial conditions and ‘animal spirits’ in the context of still high inflation, and very low unemployment.