It is a great pleasure to be able to support the Anika Foundation by speaking today. I was delighted when Adrian asked me to continue the RBA's support of the Foundation. In the early 1990s, when I first returned from studying overseas, Adrian had just taken up the position of Head of Economic Research at the RBA and he was a great mentor to me. More importantly, the Anika Foundation is undertaking really important work in helping improve the mental health of our children.

Thank you all for coming out today to support this work. I would particularly like to thank NAB for its support as the new sponsor of this lunch.

Over the past few months I have participated in a lot of international meetings: the IMF, the G20, the BIS, the Financial Stability Board and meetings of Asian central bank governors. Two points, in particular, have struck me.

The first is that conditions in the global economy have improved. The pessimism of earlier years has given way to cautious optimism. Forecasts of future growth are being revised up, not down. So the discussions are quite different from those that were taking place a year ago.

The second point is the commonality of the questions being asked about labour markets. Employment growth has generally surprised on the upside and, in a number of countries, the unemployment rate is at, or below, the rate conventionally associated with full employment. Yet at the same time, growth in wages remains subdued, even in countries with low unemployment rates. So a common question being asked is: why are the stronger labour markets not generating more upward pressure on wages? A related question is what does this mean for the outlook for inflation and monetary policy?

In today's remarks, I would like to talk about these labour market issues in an Australian context. My remarks will be in three parts. I will first talk about trends in employment in Australia. I will then discuss recent wage outcomes. And finally, the implications for monetary policy.

Trends in Employment

On a number of measures the Australian labour market has performed well over recent times. Over the past decade and a half, the unemployment rate has moved up and down within a narrow range of around 2 percentage points and its average level has been considerably below that in the previous 30 years (Graph 1). Most other countries have seen considerably larger swings in their unemployment rates over recent times. The relative stability in Australia's unemployment rate is despite us experiencing the biggest terms of trade and mining investment booms in a century. It is a considerable achievement and a testament to the flexibility of the Australian economy.

If we look at just the past few months, there has been a welcome pick-up in employment growth right across the country, after a period of softness. The forward-looking indicators suggest that employment growth will continue. Job ads, job vacancies and hiring intentions have all lifted. Businesses are also reporting better conditions than they have for some years. This is good news, particularly given that the unemployment rate is still around ½ a percentage point above estimates of full employment in Australia.^[1]

Notwithstanding these outcomes, there are other labour market developments that are causing concern in the community. There is a degree of underemployment, wage growth is slow and job security is an issue for more people. The nature of work is also changing, in both the way we work and the industries in which we work.

Over the medium term, two trends in particular stand out. The first is the growth of part-time employment. And the second is the growth of employment in the services sector.

The number of Australians working part time has grown rapidly (Graph 2). Since the 1960s, the share of part-time workers has increased threefold to nearly one-third of total employment, with the pace of this shift picking up over recent years. Since 2013, growth in part-time employment has averaged 3 per cent per year, while growth in full-time employment has averaged less than 1 per cent.

This shift in working patterns reflects both supply and demand factors.

On the supply side, many people want to work part time. Indeed, in many workplaces, employees have long been asking for greater flexibility, including the ability to work fewer hours. Part-time work leaves time for other activities, including education, caring for others and leisure.

Some insight into why people work part time can be gained from the HILDA Survey, which asks people for the main reason that they work part time (Graph 3). The most common response is that they are studying. Other frequent responses are that they are caring for children or that they simply prefer part-time work. For many people, part-time work is what they want. The fact that we have been able to accommodate this desire is a positive feature of our labour market.

There are demand factors at work as well. Many businesses benefit from having employees who work part time. But there is an element to the demand side that is not so positive. Some people are working part-time because they can't find a full-time job and others are working part-time because of job requirements. While most part-time workers are not seeking full-time employment, around one-quarter want to work more hours than they currently do. On average, they are looking to work an additional 14 hours per week, although many are not taking active steps to secure those additional hours.

So many people want to work part time, but some of these would like more hours than they currently have. This represents an additional source of unused capacity in our labour market that is not reflected in the unemployment rate. Given this, as part-time employment has grown, the RBA has paid additional attention to alternative measures of labour market slack. One measure that has conceptual appeal is an hours-based underutilisation rate, which measures the additional hours sought by workers (including those currently unemployed) relative to the total number of hours that workers would like to work. This measure shows the same general pattern as the unemployment rate, although the gap has tended to widen gradually over time (Graph 4).^[2]

Now turning to the second longer-term trend shift we have seen in our labour market – the shift to employment in the services sector (Graph 5).

Today, almost 80 per cent of Australians work in service industries, broadly defined. By way of contrast, in the 1950s only around 50 per cent of employed Australians worked in the services sector. In the past, it was common to have a full-time job producing goods. In our more modern economy, this is no longer the case.

In an effort to better understand the growth in services-sector employment, one of the exercises that we have done at the RBA is to classify around 300 individual service-sector occupations into jobs that pay hourly wages that are below average, around average and above average.^[3] We have then tracked employment growth for each of these three groups since 2000 (Graph 6).

The picture is pretty clear. The growth in service-sector employment has been strongest in those occupations with above-average rates of pay. Since 2000, over a million new higher-paying jobs have been created in the services sector. Some of the occupations where there have been large gains in employment are: medical professionals, IT managers, project administrators and sales managers. There has also been strong employment growth in occupations with lower rates of pay.

We have also conducted the same analysis for the business services and household services sectors separately (Graph 7). The growth of higher-paying jobs has been much more pronounced in business services, than it has been in household services. For the household services sector, the growth has been strongest in jobs with below-average wages. Some of the occupations where there has been a big increase in employment here include: baristas and waiters, childcare workers and aged-care workers. So it's a mixed picture.

I am often asked where future jobs growth will come from. The short answer is that it will come mainly from where it has come from in the recent past – from the myriad of occupations in the services sector. Some of these jobs will attract relatively low rates of pay, but, if our experience is a useful guide, more of these jobs will be higher-paying high-skill jobs.

With most of us working in the services sector, it's in our national interest to lift productivity growth in these industries and to develop more higher-paid high-skill jobs. Technology is important here but so too is investment in human capital. It seems probable that the next wave of growth in Australia will be driven by us building on our expertise in services. This requires investment, including in human capital.

Growth in Wages

I would now like to turn to another element of the labour market story – the slow growth in wages.

Over the past year, the Wage Price Index has increased by 1.9 per cent. This is the slowest rate of increase since this series commenced in 1997. The same picture is evident in the broader measure of average hourly earnings from the national accounts. To help see the longer-term trends, Graph 8 shows average annual growth in hourly earnings over a rolling four-year period. The most recent observation is the lowest for many decades. From the mid 1990s until a few years ago, Australians got used to average hourly earnings increasing by around 4 per cent a year. Over recent years, growth has been a bit less than half of this.

These lower wage increases have persisted for some time now. One consequence of this is that there has been a decline in expectations of future income growth. This decline is seen as more than just temporary. It is one of the factors that has been weighing on consumption growth over recent times. As households have revised down their expectations of future income growth, they have adjusted their spending too. A downward revision to expectations of income growth also means debt obligations stay higher for longer than was originally expected.

Why is this happening, both here and elsewhere in the world?

There is no single answer.

Part of the story in Australia is that our labour market has some spare capacity and we are unwinding some of the effects on wages of the mining investment boom. But this isn't the whole story, and neither spare capacity nor a mining boom explain low wages growth in some other advanced economies.

Another part of the story, particularly overseas, is slower productivity growth. In the United States, for example, low growth in wages is being matched with low productivity growth. In Australia, productivity growth has also slowed somewhat. Here, however, the slowing in earnings growth has been more pronounced than that in productivity. The result has been a decline in labour's share of national income.

None of these reasons alone appears sufficient to explain the weakness in wage growth. This suggests that there is something else going on, and that it has a global dimension.

Many workers in advanced economies feel like they face more competition. A basic principle of economics is that when you face more competition, you are less inclined to put your price, or as a worker, your wage, up.

This perception of greater competition is coming from two sources. The first is globalisation. One of the positives of globalisation is that it increases the size of market that a firm can tap. At the same time, though, globalisation increases the number of competitors that can tap your market; it increases competition. The second source is changes in technology. In some industries, advances in technology have led workers to worry about the competition from robots. At the same time, advances in technology have made more areas of the economy subject to international competition; there are fewer truly non-traded industries any more.

Perhaps as a consequence of this extra competition – or perhaps as a consequence of other forces within our societies – many workers in advanced economies feel that the world is less secure – less secure economically and less secure politically. This means that security is valued more highly. With a greater premium on security, it's plausible that workers are less inclined to take a risk by seeking larger wage increases.

One related aspect of the current labour market is a decline in job mobility. Data published by the ABS suggest that the share of employed people changing employers is around the lowest in recent decades (Graph 9). It is likely that in an environment of less job security, fewer people are inclined to switch employers. There is also a demand-side effect, with fewer firms attempting to attract workers from other firms. This is consistent with subdued wage growth.

When I spoke about this set of issues in a recent panel at the Crawford School at ANU, I made the point that some pick-up in aggregate wage growth over time would be a welcome development. Some commentators saw this as the Reserve Bank Governor making a rather unusual call to arms: a call for workers to demand larger wage increases from their employers. My intention was less dramatic.

It was simply to make the point that a gradual pick-up in aggregate wages growth would be a positive development. The best outcome for both workers and firms is for any pick-up to be underpinned by a lift in productivity growth and more high-skill jobs. But even the current rate of productivity growth could sustain some increase in wages growth over time. Indeed, some pick-up is incorporated into the Bank's forecasts for the economy. A gradual lift in wage growth is a central element in our forecast for inflation to return to around the mid-point of the medium-term target range.

Policy Implications

I would now like to turn to some of the implications for monetary policy, both globally and in Australia.

The persistent slow growth in wages is creating a challenge for central banks. It is contributing to an extended period of inflation below target. In years gone by, the more standard challenge was to keep wage growth in check, so as to stop upward pressure on inflation, which could lead to restrictive monetary policy. No advanced economy faces this challenge at present.

It is possible that things could change in the not too distant future, particularly in those countries at, or near, full employment. It may be that the lags are just a bit longer than usual. If so, we could hit a point at which workers, having had only modest pay increases for a run of years, decide that it is time for a catch-up. If such a tipping point were reached, inflation pressures could emerge quite quickly. In this scenario we could see a period of turbulence in financial markets, given that markets are pricing in little risk of future inflation.

This scenario can't be completely discounted. It would seem, though, to have a fairly low probability in Australia, especially in light of the continuing spare capacity in our labour market. The more likely case here is that wage growth picks up gradually as the demand for labour strengthens.

Globally, an alternative scenario is that the period of slow wage growth turns out to be much more persistent, partly for the reasons that I discussed earlier. In this scenario, wages growth eventually picks up, but it takes quite a while longer. If so, inflation stays low for longer, although there are other factors that could push inflation higher.

This scenario is one in which the Phillips Curve is flatter than it once was. It is one in which inflation is harder to generate. We can't yet tell though whether the Phillips Curve in Australia has become flatter, given that we have experienced relatively little variation in the unemployment rate over recent times.

The combination of a flatter Phillips Curve and inflation below target raises a challenge for central banks: how hard to press to get inflation up? For a central bank with a single objective of inflation, the answer is relatively straightforward. Inflation is too low, so you do what you can to get inflation up. If inflation doesn't increase, you need more monetary stimulus.

This approach does carry risks, though. A flatter Phillips Curve means that the monetary stimulus has relatively little effect on inflation, at least for a while. At the same time, however, the monetary stimulus is likely to push asset prices higher and encourage more borrowing. Faced with low inflation, low unemployment and low interest rates, investors are likely to find it attractive to borrow money to buy assets. This poses a medium-term risk to financial stability.

Australia's monetary policy framework is better placed to deal with this world than some others. We have a flexible medium-term inflation target that allows financial stability considerations to be taken into account in the setting of monetary policy.

Over recent times you would have noticed that we have been paying close attention to the risks in household balance sheets. Household debt is high and rising faster than the unusually slow growth in incomes.

These developments have had a bearing on the setting of monetary policy. We have not sought to stimulate a rapid lift in inflation. The fact that the labour market has been generating sufficient jobs to keep the unemployment rate broadly steady has allowed us to be patient. Our judgement has been that seeking a more rapid pick-up in inflation through yet further monetary stimulus was likely to add to the medium-term risks. Our central scenario remains for underlying inflation to pick up gradually as the economy strengthens.

Elsewhere in the world, some central banks are now starting to increase interest rates and others are considering when to withdraw some of the monetary stimulus that has been put in place. This has no automatic implications for monetary policy in Australia. These central banks lowered their interest rates to zero and also expanded their balance sheets greatly. We did not go down this route. Just as we did not move in lockstep with other central banks when the monetary stimulus was being delivered, we don't need to move in lockstep as some of this stimulus is removed.

Our decisions will continue to be made within the framework of our medium-term inflation target. We are intent on delivering Australians an average rate of inflation over time of between 2 and 3 per cent. We are seeking to do this in a way that supports sustainable growth in the economy and that best serves the public interest. To do this we need to understand developments in Australia's labour market and to take account of our decisions on balance sheets in the economy.

Thank you for listening. I look forward to your questions.

Endnotes

I would like to thank Natasha Cassidy and James Foster for assistance in the preparation of this talk. [*]

See Cusbert, T (2017), Estimating the NAIRU and the Unemployment Gap, RBA Bulletin, June, pp 13-22 [1]

This measure includes all part-time workers seeking additional hours, regardless of whether they are taking active steps to obtain the additional hours. See RBA (2017), ‘Box B: Underemployment and Labour Market Spare Capacity’, Statement of Monetary Policy, February, pp 38–40. [2]

Occupations in the business and household services sectors have been included in this analysis. [3]