Dr Luci Ellis: Full employment rate needed to be defined

Luci Ellis. Picture: NCA NewsWIRE, Monique Harmer

This is the text of the speech Westpac chief economist Dr Luci Ellis delivered at The Australian‘s Economic and Social Outlook Conference in Melbourne today.

---

The last time the unemployment rate was around its current levels, I was in preschool. Underemployment has picked up, but it is still low.

Compared with the unemployment rates of the 1980s and 1990s, when rates above 8% were common and 6% seemed as much as we could hope for, this is a remarkable achievement.

As policymakers try to tame high inflation, the question arises: can Australia hold onto the gains we have made in getting unemployment down? And can we achieve full employment?

If you don’t know what you are trying to achieve, you’ll never get there. At least now, with the White Paper, we have a definition of full employment that seems to have broad agreement. The definition given is, paraphrased, the state of the world where everyone who wants a job can get one without having to look for too long, and where the job is secure and paid decently.

There’s a lot to unpack in that definition. It captures the quality of the employment experience and whether people can get the hours of work they want. But the hours of work people want is itself partly determined by the hourly wage they can get, as well as their family and other obligations.

Westpac chief economist Luci Ellis. Picture: John Feder

This differs from economists’ definition of a NAIRU, which is simply the rate of unemployment below which inflation starts to pick up. Or more directly, and to my mind preferably, the rate of unemployment below which wages growth starts to pick up – the NAWGRU, if you like.

I prefer a definition keying off wages growth because, as we have seen recently, inflation in goods and services prices can be affected by factors beyond the cost of labour. If you’re not careful, you would end up attributing inflation that stemmed from supply shocks to a rising NAIRU. Indeed, if you were to run the standard models now, with no account for the pandemic, they would see the global inflation surge and conclude that the NAIRU was currently around 6%. I don’t believe that for a minute.

So full employment and the NAIRU are two different things. The concept of full employment is normative, a goal we have. While the concept of NAIRU is empirical. It is how things are.

If the NAIRU/NAWGRU accords with full employment, then the aspiration can be sustained. But if not, we can’t stay at full employment for long without other problems arising.

If the NAIRU is above the aspiration for full employment, this does not mean we have to give up. The NAIRU is not fixed or handed down from the heavens. It is an emergent property of a complex system. It absolutely can shift.

One way, therefore, to think about full employment as a policy objective is:

• First, to find a way to ensure that the NAIRU declines to be close to full employment.
• And second, to make sure that policymakers either know this has happened or at least conduct policy in a way that allows for the possibility. Without this, macro policy could end up being too tight until the policymaker learns from experience that the NAIRU has fallen.

This was possibly part of the post-GFC story. In the late 2010s, many countries found they could get unemployment lower than previously, without igniting inflation. It is no surprise that during this period, official estimates of NAIRUs for advanced economies, such as those produced by the OECD, were repeatedly revised down.

The first part of the objective – reducing the current level of the NAIRU from wherever it is – is usually framed as being an issue with the worker. The policy focus is all about removing barriers to employment, improving skills, lifting literacy and so on. And these things are worth doing. But I question whether they are the whole story.

For no matter how job-ready the potential worker might be, there has to be a firm ready with a job. As well the firm having work that needs doing, the firm needs to see that worker as suitable for the role. This comes down to employer practice as much as the worker’s actual characteristics. Firms have already seen the benefit of increasing the diversity of their workforce across dimensions such as sex, age and cultural background. But there can still be a bit of a tendency to want to hire people into roles who have already done that role elsewhere.

When the labour market isn’t as tight as it is now, it is common for workers to complain that job selection criteria don’t always make sense. That you need a qualification that the nature of the work doesn’t really require. That you need five years’ experience for an entry-level position, or in a technology platform that has only been around for three years.

As the labour market tightens up, firms do learn not to be so specific. They start taking on school leavers or new graduates, or people with a non-standard collection of experience. After all, in a few years’ time, the inexperienced person becomes an experienced person.

But it takes time for firms to learn that this is what they need to do, that isn’t a temporary situation they can ride out. And no doubt it takes the HR department a while to agree with the changed approach, organisations being what they are.

This isn’t a critique of employers or HR professionals by any means. It is simply the fact that any single person or organisation has only one window into the whole labour market. So it isn’t always obvious to people whether the market has tightened, or they’ve just been unlucky in their search.

We economists have a fancy word for this process of lower unemployment nudging the NAIRU lower – hysteresis. Normally we think about the reverse case, though, where high actual unemployment drags the NAIRU up. But hysteresis works both ways, and this search and learning process is probably a lot of the reason for it. It means that if we find that the economy can run with a lower average unemployment rate without igniting much wages growth, then testing that boundary just a little can result in better outcomes over time.

So the first part of the problem, getting the NAIRU down, has three elements:

1. All the usual worker-centric policies around training, removal of barriers and job-readiness
2. Running the economy a little tight, but not too tight, so that firms have the incentive to learn that they can hire someone who isn’t already doing the same job somewhere else, perhaps with some training. Of course, this cannot go too far, else inflation will increase. It is an argument for approaching the limits slowly, to create that learning time, rather than pushing for the fastest possible return to wherever the NAIRU is estimated to be now.
3. Related to this, finding ways to speed up firms’ learning process, so that it isn’t the case that every episode of labour market tightening is construed as a labour shortage. I don’t pretend to have all the answers on how to make this happen, but this seems like the right forum to ask the question. A good place to start would be to ensure that the process of matching firms and workers is as efficient as it could be. The private sector has a role to play here, through platforms and other initiatives. It isn’t the sole responsibility of government.

The second part – making sure policymakers realise that the NAIRU can decline or has declined – is a task for the economics profession. I would like to encourage my academic colleagues to explore the underlying drivers of this emergent property of the system. It would also be worth investigating whether it is feasible to forecast its movements. The ageing of the population and rising participation of older workers will be a factor here.

The profession can also assist in keeping policymakers alert to these issues. The good news is that, in Australia at least, the culture of public policy is pragmatic and evidence based. I do not see too many people in that sphere who take models and invisible “star” variables as gospel. So I do not see much risk here of a policy mistake coming from decisions that assume a fixed, and too-high, NAIRU. And for that reason, I believe that full employment and low inflation are compatible, at least in principle, and with time.

To do that, we need to make sure that we aren’t caught up in preconceived ideas. We can do more to recognise that the NAIRU can move, that it might be able to be forecast, and that – perhaps – we can do more to influence it in the right direction, so that our full employment aspirations can be achieved.

And by “we”, I don’t only mean government policy. As with Australia’s recent weak productivity outcomes, it is too often assumed that causes and solutions rest entirely on government actions. Productivity isn’t something the government does to us, though. It is mostly the result of the decisions every business makes: how much to invest; when and how to train employees; and how work is organised and prioritised.

So it is with the emergent properties of the labour market. Public policy has an influence, but it isn’t the only thing. Private sector decisions matter, including who and how we hire, how we on-board and train staff, and how we organise our work and – especially – our working hours and locations. Tighter labour markets require nimbleness and creativity in filling roles, but these skills can atrophy in times of labour market slack.

At a time when participation in the labour market is close to an 113-year high, we can’t say that there is a shortage of willing labour. The challenge is to match that willing labour with suitable roles to achieve our aspirations of full employment. That is something we can all participate in, rather than waiting for the government to do it for us.