One of the biggest economic mysteries since the end of the GFC has been the lack of wages growth globally, perplexing policy makers for whom it has become a key indicator of monetary policy. We may need more than low unemployment to generate wages growth or inflation and given growing inequality globally, it might be time for governments to step up with alternative policies.

That has been highlighted by many economists, central bankers and policy makers, most recently speeches from Reserve Bank of Australia (RBA) Governor Phil Lowe on 26 July and the Bank of England’s (BoE) Chief Economist, Andy Haldane on 20 June focused on this issue as a key topic. 

Globalisation, technology are breaking down historical wages & productivity relationship

When considering wages growth it is important to consider that wages growth should be fundamentally driven by productivity growth, and indeed a decline in productivity growth can explain some of the decline in wages growth, particularly in the US. But it does not explain all the variation. 

Another factor that some, including Governor Lowe have pointed to is increased globalisation and improving technology. Not only are markets more open to global competition but technology is making more products (particularly services) tradable, opening them up to competition. Due to this increased competition it’s not just workers in manufacturing that are being disrupted but services too. Digital or online services in particular are being disrupted, it is now cheaper (and sometimes easier) to hire a graphic designer from the Ukraine rather than Australia, similarly customer service (particularly online and telemarketing) has seen much outsourcing in recent years.  While these sectors are the obvious starting point we will likely see this trend broaden into others as technology continues to make the world a smaller place.

Shift towards workplace flexibility: “it’s not you….it’s me…..”

Probably the largest driver of this disconnect though is one highlighted by the BoE’s Andy Haldane; the nature of work is changing and this is leading to shifting relationships between employers and employees. This includes the decline of collective bargaining partly due to lower union membership, the shift towards self-employment (the “gig” economy), the rise of part-time work and the increasing use of zero hour contracts. What all these changes do is make work more divisible (at both the task and worker level) and to a certain extent more flexible, with more employees being paid on an hourly or task based basis.

According to the BoE the share of the workforce in the UK either self-employed, part-time, temporary or working on zero-hour contacts has risen to around 43%, up from 39% in 2000. While this is not a majority and sounds like a small increase it represents around 3 million workers. Importantly this also means the marginal worker is now much more likely to be self-employed or work flexibly and as we know from economics, the marginal worker will set wages.

Not everyone is a fan of the new arrangements 

While this increase in flexibility has been positive and desired for many and brought more women in particular into the workforce (in Australia the three most common reasons for people working part-time are study, preference and caring for children), there is also a significant group for whom it is not a choice and who would like to work more hours. 

Measures of underemployment or underutilisation can help show this story. While unemployment has fallen to pre-crisis levels in the US, UK, Europe and Japan, underemployment/underutilisation rates remain high, likely for structural reasons. Indeed using these measures instead of unemployment makes the Phillips curves look more traditional, but still flatter than in the past.

History suggests, the Phillips curve will remain flat

Looking at history it is interesting to consider that if anything, we are not moving to a new way of working but returning to an old one. As Andy Haldane notes “prior to the Industrial Revolution… most workers were self-employed or worked in small businesses… hours were flexible, depending on what work was needed… [and] work was artisanal, task-based [and] divisible.” While the comparison is clearly far from perfect it is interesting to consider the historic experience in relation to wage growth. The chart below shows the Phillips curve for the UK for four periods: pre-industrial, post-industrial, the post-war years and the late 70’s to today. 

The Phillips curve, working the angles:  

Source: Thomas, R and Dimsdale, N (2017), ‘A millennium of macroeconomic data’, Bank of England OBRA dataset and Bank of England calculations.

As the chart illustrates, the Phillips curve for the pre-industrial period is almost flat and bears a close resemblance to Phillips curves since 2008. While this is far from conclusive evidence it does support the idea that a shift in work practices and the changing nature of employment have contributed to a flattening of the Phillips curve. With little sign of these structural changes reversing it seems we may be returning to a (relatively) flat Phillips curve world.

All this suggests that we may likely need more than just low unemployment to generate wages growth or inflation and given growing inequality globally, it might be time for governments to step up with alternative policies. It also supports a recent trend of central banks considering other factors, besides inflation, when setting policy, such as the aggregate level of growth and financial stability concerns. Which begs the questions, given how much faith Central Bankers and policy makers have placed in the Phillips Curve, when is it time to call it dead?