Topic 5 of 8

Jobs & the economy

How unemployment is measured, how the RBA links it to inflation through the NAIRU, and the MMT case for a Job Guarantee, with its critics.

5 lessons · about 1 h 5 min with activities, 2 h 10 min with the long lectures · beginner first

Before the videos

Key ideas for this topic

The primer explains how the ABS counts jobs and unemployment and how the RBA links unemployment to inflation, followed by a short written lesson on the NAIRU logic. A 57-second news clip then tests that logic. Bill Mitchell's two videos argue that unemployment is a policy choice and explain the Job Guarantee, and Pavlina Tcherneva's hour-long lecture is optional depth.

Builds on: The Inflation topic showed how the RBA uses interest rates to bring inflation down; this topic looks at what that means for jobs, and who bears the cost.

Ideas you'll need

1Who counts as unemployed?

Every month the ABS surveys about 24,000 households, roughly 50,000 people, for its Labour Force survey. Each person aged 15 and over is placed in one of three groups, based on a single reference week.

Employed: did at least one hour of paid work. A student with a four-hour Saturday shift counts.
Unemployed: had no paid work, had actively looked for work at some point in the past four weeks, and was available to start. Simply wanting a job is not enough. People waiting to start a job within four weeks also count as unemployed, and people with a job who were away from it, for example on leave, count as employed.
Not in the labour force: everyone else, such as retirees, many full-time students, and carers who are not looking for work.

The labour force is the employed plus the unemployed. The unemployment rate is the number of unemployed people as a share of the labour force. In August 2026 it was 4.6 per cent, about 723,000 people. The participation rate, the labour force as a share of everyone aged 15 and over, was 67.1 per cent.

The one-hour rule surprises people. It means the headline unemployment rate is a narrow measure of how much work people want but cannot get, which is why the next idea matters.

2Underemployment and the kinds of unemployment

Someone working 10 hours a week who wants 30 counts as employed. The ABS also counts them as underemployed: employed people who want, and are available for, more hours. In August 2026 the underemployment rate was 6.2 per cent, higher than the unemployment rate. Together they give a broader picture of unused labour.

The RBA separates three kinds of unemployment:

Frictional: the short gap while people move between jobs. Some is normal.
Structural: a mismatch between the jobs on offer and the skills or locations of the people looking. A retrenched Hunter Valley mine worker and a nursing vacancy in Sydney do not match. It can last more than a year, even in good times.
Cyclical: unemployment that rises when total spending in the economy falls, as in a recession, and falls when spending recovers.

The split matters for policy. The RBA says interest rates mainly affect cyclical unemployment, while skills, training and job-matching policies deal with structural unemployment. MMT economists such as Bill Mitchell argue that much of what gets labelled structural is really a long-running shortage of demand.

3The Phillips curve and the NAIRU

Economists noticed that inflation tended to be higher when unemployment was low, and lower when it was high. A graph of that relationship is called a Phillips curve. The idea behind it: when unemployment is very low, businesses compete for scarce workers, wages rise faster, and businesses pass the extra costs into prices.

The RBA describes a short-run trade-off: lower interest rates can cut unemployment, at the cost of higher inflation. In the long run, it says, the trade-off disappears. Unemployment returns to a level set by the structure of the economy, and trying to hold it lower only raises inflation.

That level is the NAIRU, the non-accelerating inflation rate of unemployment. The RBA defines it as the lowest unemployment rate that can be sustained without wages growth and inflation rising. Below it, the RBA expects inflation pressure to build; above it, there is spare capacity.

Two cautions. The NAIRU cannot be observed, only estimated: a 2017 RBA estimate of about 5 per cent came with a band of roughly 4 to 6 per cent, and estimates have fallen since: in February 2024 the RBA's models put the unemployment rate consistent with full employment at about 4 to 5½ per cent. Unemployment of 4.6 per cent in August 2026 sits inside that range, not clearly above it. And long spells of unemployment can erode skills and push the NAIRU itself up, an effect economists call hysteresis.

4Theory, objective, evidence: keep them apart

Arguments about jobs often blur three different kinds of claim.

A theory of unemployment explains what causes it. The NAIRU framework says unemployment settles around a level set by the economy's structure, with spending moving it around that level. MMT and post-Keynesian economists put more weight on a shortage of total spending. Theories can only be tested indirectly.
A policy objective says what governments should aim for. The RBA's mandate is low, stable inflation (2 to 3 per cent) and full employment, which it describes as maximising employment while keeping inflation low and stable. The government's 2023 Employment White Paper lists 'sustained and inclusive full employment' as its first objective. Objectives involve value judgements.
An empirical claim is about what the data show: unemployment averaged under 2 per cent in the 1960s; the RBA estimated the NAIRU at about 5 per cent in 2017.

Two people can share an objective, such as full employment, and disagree about the theory. Others can accept the same data and weigh the costs differently. As you watch each video, ask which kind of claim the speaker is making.

Saved in this browser only. It records that you marked it, nothing more.

Lesson 5.1 Beginner6 min read Core about 12 min with questions

How the RBA thinks about unemployment and inflation

A written lesson from Progressive Money

Before you read

After reading this, you should be able to explain why the RBA expects unemployment to rise when it raises interest rates, and what its framework does not claim.

Why this matters. Every video in this topic argues with this framework, for or against. Knowing what the RBA actually says lets you judge the critics fairly, including the ones you agree with.

Start with a labour market that is running hot. Unemployment is very low, businesses say they cannot find staff, and to hire or keep people they offer bigger pay rises. If pay grows much faster than productivity, businesses pass some of the extra cost on in prices. Households are also spending strongly. Inflation sits well above the RBA's 2 to 3 per cent target.

What the RBA does. The Monetary Policy Board raises the cash rate. Mortgage repayments go up, households have less to spend, and businesses put off investment. As spending slows, businesses need fewer extra workers. Job ads fall, hiring slows, and the unemployment rate drifts up. With less competition for workers, wage growth eases, and so does inflation.

Where the NAIRU comes in. The RBA describes the NAIRU as the lowest unemployment rate that can be sustained without wages growth and inflation rising. If unemployment looks to be below it, the RBA reads the labour market as tight and expects inflation pressure. If unemployment is above it, there is spare capacity and the RBA can cut rates. The NAIRU cannot be observed. A 2017 RBA estimate put it at about 5 per cent, with a band of roughly 4 to 6 per cent; later research put it below 5 per cent before the pandemic. In February 2024 the RBA's models put the unemployment rate consistent with full employment at about 4 to 5½ per cent.

A real example. On 5 September 2023, after a long run of rate rises, the RBA said it expected unemployment to rise gradually to around 4½ per cent by late 2024, while inflation returned to the target range in late 2025.

What the framework says, and what it does not.

It does not aim for unemployment as a goal. The RBA's mandate is low, stable inflation and full employment, which it describes as maximising employment while keeping inflation low and stable.
It does accept that unemployment may rise as a side effect of bringing inflation down. Its reasoning is that, over the long term, low and stable inflation supports sustainable employment.
It holds that interest rates can only move unemployment around the NAIRU in the short run. Lowering the NAIRU itself depends on other government policies, such as skills, training and job matching.

Why critics object. MMT economists such as Bill Mitchell and Pavlina Tcherneva argue that this framework uses unemployed people as the economy's inflation buffer, that NAIRU estimates are too unreliable to steer by, and that a better buffer exists. Mainstream economists reply that the uncertainty is real but the link between a tight labour market and inflation is not imaginary. The rest of this topic weighs those claims.

Three things to remember

Established The NAIRU cannot be observed, only estimated. A 2017 RBA estimate of about 5 per cent came with a 70 per cent confidence band of plus or minus one percentage point, and estimates have drifted down since, to about 4 to 5½ per cent across the RBA's models in early 2024.
Official view The RBA's mandate is 2 to 3 per cent inflation and full employment. When it raises rates to cut inflation it expects unemployment to rise somewhat, as in its September 2023 forecast of around 4½ per cent.
Argument MMT economists such as Bill Mitchell and Pavlina Tcherneva argue the NAIRU framework treats unemployed people as an inflation buffer, and that its estimates are too uncertain to guide policy.

Can you explain it?

Optional. Try as often as you like; nothing is scored or sent anywhere.

Unemployment is 3.5 per cent, inflation is 6 per cent, and the RBA raises the cash rate. Which statement best describes its framework?

It expects slower spending to ease hiring and wage pressure, and accepts unemployment may rise somewhat as inflation falls. It is trying to raise unemployment for its own sake. It believes unemployment and inflation are unrelated. It has calculated exactly how high unemployment needs to go.

Show the answer

It expects slower spending to ease hiring and wage pressure, and accepts unemployment may rise somewhat as inflation falls.

Yes. Higher unemployment is an expected side effect of slowing total spending, not the aim. The RBA's mandate includes full employment alongside low inflation.

Which of these is a policy objective, rather than a theory of unemployment or an empirical claim?

Monetary policy should aim for full employment and 2 to 3 per cent inflation. Holding unemployment below the NAIRU makes inflation keep rising. Unemployment averaged under 2 per cent in the 1960s. The RBA estimated the NAIRU at about 5 per cent in 2017.

Show the answer

Monetary policy should aim for full employment and 2 to 3 per cent inflation.

Yes. This says what policy should aim for. It is written into the RBA's mandate, and choosing it involves value judgements about what matters.

Sources for this lesson

Lesson 5.2 Beginner1 min video Core about 7 min with questions

Property developer says unemployment should rise

SBS News

Before you watch

After watching this video, you should be able to explain what the RBA's framework does and does not imply about rising unemployment, using Gurner's comments as a test case.

Why this matters. The clip went viral because many viewers heard it as the system's hidden logic. Separating what the RBA says from what Gurner said is a good test of the previous lesson.

Play: Property developer says unemployment should rise (1 minutes, opens YouTube)

A 57-second SBS News clip from September 2023. Property developer Tim Gurner told a property industry summit that unemployment needed to rise 40 to 50 per cent because workers had become 'arrogant', and apologised within days. We use it as a discussion piece: having just seen how the RBA thinks about unemployment and inflation, ask how much of his claim that framework actually supports.

Three things to remember

Established In September 2023 Gurner said unemployment needed to rise 40 to 50 per cent to curb worker 'arrogance'. After criticism from the AMA and MPs across parties, he apologised and said his comments were wrong.
Official view A week earlier the RBA forecast that unemployment would rise gradually to around 4½ per cent as inflation returned to its 2 to 3 per cent target. It presented this as easing a tight labour market, not as a goal.
Argument Gurner framed unemployment as a way to discipline workers. That is his value judgement. The RBA's reasoning is about total spending and wage pressure, and no mainstream institution endorsed his framing.

Can you explain it?

Optional. Try as often as you like; nothing is scored or sent anywhere.

The unemployment rate was 3.7 per cent in August 2023. If it rose by 50 per cent, the top of Gurner's range, what would the unemployment rate be?

Your answer

Show the working

Answer: 5.55%

A 50 per cent rise multiplies the rate by 1.5: 3.7 × 1.5 = 5.55 per cent. It is not 53.7 per cent, because a rise of 50 per cent is not a rise of 50 percentage points. About 540,000 people were unemployed then, so this would mean roughly 270,000 more people out of work.

Which statement is closest to what the RBA's framework implied about Gurner's claim in September 2023?

The RBA expected some rise in unemployment as it slowed spending, but its aim was low inflation and full employment, not disciplining workers. The RBA also wanted unemployment to rise 40 to 50 per cent. The RBA's framework says unemployment has no effect on inflation, so Gurner was simply wrong. The RBA agreed workers had become too demanding and needed to be brought into line.

Show the answer

The RBA expected some rise in unemployment as it slowed spending, but its aim was low inflation and full employment, not disciplining workers.

Yes. The RBA forecast a modest rise as a side effect of bringing inflation down. Gurner's language about 'arrogance' and the need for 'pain' was his own framing.

Sources for this lesson

Lesson 5.3 Beginner19 min video Go deeper about 27 min with questions

Professor Bill Mitchell - The Government Chooses the Unemployment Rate

Bill Mitchell

Before you watch

After watching this video, you should be able to explain what Mitchell means by 'the government chooses the unemployment rate' and the main mainstream objection to it.

Why this matters. Australia once had unemployment under 2 per cent for years. Whether that was a policy choice we could repeat, or a product of its time, is central to today's debate about full employment.

Play: Professor Bill Mitchell - The Government Chooses the Unemployment Rate (19 minutes, opens YouTube)

An 18-minute talk Mitchell gave at the Full Employment Workshop in Sydney on 30 May 2015, the 70th anniversary of Australia's 1945 White Paper on Full Employment. He traces the postwar decades of very low unemployment, argues they ended with a political shift in the 1970s, and makes his central claim: the federal government chooses the unemployment rate. Treat that as an argument to test, not a settled finding.

Three things to remember

Established Australian unemployment averaged under 2 per cent in the 1960s, then rose sharply: from 1.9 per cent in 1972 to 9.4 per cent in 1983, according to Treasury and economist Bob Gregory.
Official view The RBA says monetary policy targets cyclical unemployment, while structural and frictional unemployment are better addressed by government policies on skills and labour supply.
Argument Mitchell argues a currency-issuing government can always afford to employ idle workers, so unemployment above a small frictional level is a policy choice, and that the postwar decades show it can be done.

Can you explain it?

Optional. Try as often as you like; nothing is scored or sent anywhere.

Mitchell says the federal government 'chooses' the unemployment rate. Which is the most accurate reading of his claim?

Because it issues the currency, it could always fund enough spending or jobs for everyone who wants work, so leaving people unemployed is a decision. Ministers literally set an unemployment target each year. Unemployment mostly reflects the choices of unemployed people. The RBA sets the unemployment rate through the cash rate alone.

Show the answer

Because it issues the currency, it could always fund enough spending or jobs for everyone who wants work, so leaving people unemployed is a decision.

Yes. That is his argument. Critics accept the financial point but dispute whether this could be done without inflation, which is where the real disagreement lies.

Find one theory, one policy objective and one empirical claim in Mitchell's talk. Which is easiest to test, and why?

Your answer (stays in this browser, never sent anywhere)

Show a model answer

Theory: unemployment happens when total spending falls short of what is needed to employ everyone who wants work. Objective: governments should guarantee full employment, as the 1945 White Paper committed them to. Empirical claim: unemployment was rarely above 2 per cent in the postwar decades. The empirical claim is easiest to test against Treasury and ABS data, and it largely holds up. The theory is harder, because mainstream economists accept that demand matters but also point to structural causes and inflation limits. The objective rests partly on values.

A strong answer

Correctly separates a cause-and-effect claim (theory) from a goal (objective) and a fact about the data.
Notes that the historical data can be checked against official statistics.
Recognises that the theory is where mainstream economists mainly disagree.

Where this is contested

Can the federal government choose the unemployment rate?

Mitchell's claim that the government 'chooses' the unemployment rate is contested. Mainstream economists agree that spending can reduce cyclical unemployment, but argue that pushing too far raises inflation and that structural unemployment needs other remedies.

Unpack the disagreement

What is established?

Australia's federal government issues the dollar, so it can always make payments in it. Unemployment averaged under 2 per cent in the 1960s and rose sharply in the 1970s and early 1980s. The RBA distinguishes cyclical unemployment, which stronger spending can reduce, from structural and frictional unemployment, which persist even in good times.

What does Mitchell argue?

Mitchell argues that unemployment above a small frictional level exists because total spending is too low to employ everyone who wants work, and a currency issuer can always close that gap. With a Job Guarantee paying a fixed wage, he says, it can do so without driving up inflation, because the government hires at a fixed price instead of bidding against private employers. The postwar decades, he argues, show this worked while governments were committed to it.

What do mainstream economists argue?

Mainstream economists agree that spending affects unemployment and that the government cannot run out of its own currency. They argue the binding limit is inflation: push demand too far and businesses compete for scarce workers, wages and prices accelerate, and policy must pull back. They also see genuine structural barriers, such as mismatches of skills and location, that spending alone does not fix. Treasury has described the 1970s rise in unemployment as a structural deterioration, not simply a change of political will.

Where is the actual disagreement?

The disagreement is about mechanisms, not accounting. Does a fixed-wage job offer escape the inflation limit that ordinary stimulus runs into? How much of today's unemployment is structural, and how much is a shortage of demand? There is also a values question: how much inflation risk is worth accepting for lower unemployment.

What evidence matters?

Question for you

In your own words, why might a Job Guarantee hiring at a fixed wage behave differently from ordinary government stimulus, and what would a sceptic say?

Your answer (stays in this browser)

Show a model answer

Ordinary stimulus raises spending across the economy, so once workers are scarce, businesses bid against each other and wages and prices rise. A Job Guarantee only hires people nobody else wants to hire, at a fixed wage, so in Mitchell's view it never bids up wages. A sceptic would say the extra income still gets spent in the wider economy, that a secure job offer may strengthen workers' bargaining power and push wages up, and that structural barriers such as skills or location do not disappear because a job is on offer.

Sources for this lesson

Lesson 5.4 Beginner8 min video Core about 14 min with questions

MMT: What Is A Job Guarantee (or "Labor Standard")?

Bill Mitchell · Deficit Owls

Before you watch

After watching this video, you should be able to explain the buffer stock idea behind the Job Guarantee and how it differs from using unemployment to restrain inflation.

Why this matters. The Job Guarantee is MMT's main answer to the NAIRU. Understanding the buffer stock logic lets you judge it on its merits rather than as a slogan, for or against.

Play: MMT: What Is A Job Guarantee (or "Labor Standard")? (8 minutes, opens YouTube)

Bill Mitchell explains the Job Guarantee in this eight-minute interview from 2014. He traces the idea to a 1978 agricultural economics class at Melbourne University and Australia's wool price scheme, under which the government bought surplus wool at a set price. Swap wool for labour and you have a buffer stock of workers employed at a fixed wage. A short foundation for Tcherneva's longer lecture.

Three things to remember

Established Australia's wool reserve price scheme, introduced in 1974, guaranteed growers a minimum price. It was suspended in 1991 after demand fell below a price set in boom years, leaving a stockpile of 4.7 million bales.
Argument Mitchell argues the government can do for labour what the wool scheme did for wool: offer a job at a fixed wage at the bottom of the pay scale, so private employers can hire workers back whenever they offer more.
Argument Mitchell argues the Job Guarantee also sizes the deficit automatically: spending rises until the last jobseeker has a job. He also argues that even work of little market value is better than unemployment, a claim critics such as Sligar and Sturgess dispute.

Can you explain it?

Optional. Try as often as you like; nothing is scored or sent anywhere.

In Mitchell's buffer stock, what plays the role that the stockpile of wool played in the wool scheme?

Workers employed in the Job Guarantee at a fixed wage. Unemployed people receiving JobSeeker. Commonwealth bonds held by investors. Unfilled job vacancies in private businesses.

Show the answer

Workers employed in the Job Guarantee at a fixed wage.

Yes. The government 'buys' labour nobody else wants at a fixed price, and private employers can buy it back by offering better terms when the economy picks up.

Suppose a Job Guarantee paid $25 an hour for a 38-hour week. What would one full-time worker's wages cost over 52 weeks, before any other costs?

Your answer

Show the working

Answer: $49,400

$25 × 38 = $950 a week, and $950 × 52 = $49,400 a year. For 100,000 workers that is about $4.9 billion in wages alone, before supervision, materials and on-costs, but partly offset by lower JobSeeker payments and extra tax. Those offsets are why cost estimates vary so widely.

Where this is contested

Would a Job Guarantee work as its proponents claim?

The Job Guarantee is MMT's most contested proposal. Critics question its cost, administration, inflation effects, the value of the work and its effect on other jobs; proponents argue it costs less than unemployment and anchors wages. Evidence from comparable programs is mixed and none is a full test.

Unpack the disagreement

What is established?

Programs with job-guarantee features exist. India runs a very large rural employment guarantee, and a rigorous study of a reform to it found beneficiary earnings rose 14 per cent, mostly through higher private wages and employment. Australia has run related schemes, such as Work for the Dole. No country has run a universal, permanent job guarantee of the kind MMT economists propose, so there is no direct test.

What do Mitchell and other proponents argue?

Cost: modest once welfare savings and extra output are counted; Levy Institute modelling of a US scheme put the net budget cost at about 1 to 1.5 per cent of GDP.
Inflation: the scheme hires at a fixed wage and never bids against private employers, so it anchors wages rather than chasing them.
Productivity: care and environmental work has social value even without a market return.
Administration: a stable core of jobs plus a smaller part that grows and shrinks with the economy, run locally.
Displacement: the wage sits at the bottom, setting a floor rather than poaching workers.

What do mainstream economists argue?

Australian critics, including Adam Triggs and, separately, David Sligar and Hugh Sturgess, raise five problems.

Cost: common Australian estimates run to $20 to 30 billion a year, though Triggs thinks this is the weakest objection.
Inflation: below the natural rate of unemployment, wages and prices still rise.
Productivity: long-term unemployed people often need health care or training, not just a job.
Administration: a large bureaucracy that must expand and shrink with the cycle.
Displacement: skilled work competes with existing jobs; basic work risks becoming make-work like Work for the Dole.

Where is the actual disagreement?

Partly empirical: does a fixed wage really anchor inflation, and how much useful work exists at scale? Partly about state capacity: could Australian governments run it well? And partly about values: is a guaranteed job better than higher income support? Evidence from real programs cuts both ways, and none is a full test.

What evidence matters?

Question for you

Pick the critique of the Job Guarantee you find strongest and explain how a proponent would answer it. Has the answer convinced you, and why?

Your answer (stays in this browser)

Show a model answer

Example: the displacement critique. If Job Guarantee workers do useful work such as aged care or bush regeneration, they may take work that would otherwise go to paid staff on award wages, undercutting them. A proponent would answer that jobs are designed to meet needs nobody currently pays for, and that the low fixed wage means employers can always hire workers away. Whether that convinces depends on whether you think a large program could be run with that discipline in practice, which is partly a judgement about state capacity.

Sources for this lesson

Lesson 5.5 Intermediate58 min video Long lecture · optional about 1 h 5 min with questions

Pavlina Tcherneva — The Case for a Job Guarantee

Pavlina Tcherneva · Watson School of International and Public Affairs

Before you watch

After watching this video, you should be able to describe the main design features of Tcherneva's Job Guarantee and explain one sceptical question she answers.

Why this matters. This is the fullest case for the Job Guarantee on the site, covering design, cost and politics. The questions from sceptical economists in the second half are as useful as the talk.

Play: Pavlina Tcherneva — The Case for a Job Guarantee (58 minutes, opens YouTube)

An hour-long online seminar at Brown University's Rhodes Center, hosted by Mark Blyth, on Tcherneva's book The Case for a Job Guarantee. She traces the right to work through US history, argues that the NAIRU has become an excuse for tolerating unemployment, and sets out her design: a permanent, voluntary public job offer at a living wage. The examples and costings are American; the questions from other economists are worth staying for.

Three things to remember

Established Large programs with job-guarantee features already exist, notably India's rural employment guarantee. A rigorous study found a reform to it raised beneficiary earnings 14 per cent, mostly via private wages.
Argument Tcherneva argues the NAIRU has become a convenient excuse for keeping some people unemployed, and that a permanent, voluntary, federally funded job offer would stabilise the economy and set a labour standard.
Argument Levy Institute modelling she co-wrote estimated a US program of up to 15 million jobs at US$15 an hour would cost a net 1.1 to 1.5 per cent of GDP and add under one point to inflation at its peak.

Can you explain it?

Optional. Try as often as you like; nothing is scored or sent anywhere.

A Brown economist asks whether the case for jobs as a right depends on the NAIRU being wrong. What is the best way to describe the two claims?

One is a values claim about rights, the other an empirical claim about inflation; they can be argued separately. Both are empirical claims that data could settle. If a NAIRU exists, a right to work is impossible. If the NAIRU does not exist, a right to work follows automatically.

Show the answer

One is a values claim about rights, the other an empirical claim about inflation; they can be argued separately.

Yes. Tcherneva largely accepts this split. You could believe in a right to work and still worry about inflation, or doubt the NAIRU and still oppose a guarantee.

Tcherneva's examples are American. Name two ways an Australian Job Guarantee would face different conditions, and say why they matter.

Your answer (stays in this browser, never sent anywhere)

Show a model answer

Australia already has a national minimum wage set by the Fair Work Commission, so a Job Guarantee wage would be less about lifting the wage floor than in Tcherneva's US case. Australia also has universal Medicare, so health cover would not be a selling point of guaranteed jobs. Australia already runs Work for the Dole, so a guarantee would be judged against that experience, including its 'lock-in' problems. And programs would need to work across federal, state and local governments and in remote communities.

A strong answer

Identifies at least two concrete institutional differences, such as minimum wage setting, Medicare or existing mutual obligation schemes.
Explains why each difference changes the costs, benefits or design of the program.

Where this is contested

Is a Job Guarantee better than the alternatives?

Tcherneva's case rests on two separate claims: a values claim that people have a right to paid work, and an empirical claim that a Job Guarantee would stabilise the economy better than unemployment does. Critics dispute the second and prefer other remedies.

Unpack the disagreement

What is established?

Australia supports unemployed people mainly through income support (JobSeeker) with mutual obligations, including Work for the Dole, plus training and wage subsidies. Long spells of unemployment are linked to poorer health, and the RBA notes they can erode skills and push up the NAIRU, an effect called hysteresis.

What does Tcherneva argue?

Tcherneva argues that only a job guarantee closes the gap income support leaves: people want work, not just income, and benefits do not create the missing jobs. As a permanent, voluntary program it would act as an automatic stabiliser, set a labour standard and prevent the social costs of unemployment. She rejects a universal basic income as a substitute, and argues that the real costs of unemployment already exceed the financial cost of the program.

What do mainstream economists argue?

Critics prefer other tools. Adam Triggs favours conventional fiscal and monetary stimulus plus tax and welfare reform. David Sligar and Hugh Sturgess propose raising JobSeeker, hiring public servants on normal terms, and a participation income that recognises caring and volunteering; they argue a Job Guarantee would work like Work for the Dole with better pay. Both sides accept unemployment is costly. They differ on the remedy.

Where is the actual disagreement?

Mostly about design and values. Is it paid work itself, or income and social standing, that protects wellbeing? Can governments create enough useful jobs without crowding out existing ones? Evidence from Work for the Dole's lock-in effects and from India's program points in different directions.

What evidence matters?

Question for you

Tcherneva and her critics agree unemployment is costly. In your own words, what do they actually disagree about?

Your answer (stays in this browser)

Show a model answer

They disagree about the best remedy, not the problem. Tcherneva thinks a guaranteed job does things income support cannot: it gives people work they want, keeps skills current and stabilises the economy. Critics think higher income support, training and ordinary public sector hiring could achieve most of that with fewer risks, such as make-work or displacing existing jobs. Underneath is a question of evidence, about whether paid work itself protects wellbeing, and a question of values, about how far governments should guarantee outcomes.

Sources for this lesson

The big debate in this topic

Do we need some unemployment to keep inflation low?

Where this is contested

Do we need some unemployment to keep inflation low?

Unpack the disagreement

What is established?

The RBA's mandate covers both full employment and low, stable inflation, and its framework assumes a short-run trade-off between the two. The NAIRU cannot be observed. RBA estimates carry a band of about plus or minus one percentage point, and have moved a lot: from just over 7 per cent in 1995 to around 5 per cent in 2017.

What do MMT economists argue?

MMT economists such as Bill Mitchell and Pavlina Tcherneva argue that NAIRU estimates are too unreliable to steer by, and that using unemployment to control inflation imposes large, lasting costs on the people least able to bear them. They propose a different buffer: a Job Guarantee that employs anyone who wants work at a fixed wage. Because it never bids against private employers, they argue, it would anchor wages and prices while keeping people employed.

What do mainstream economists argue?

The RBA and most mainstream economists accept that the NAIRU is uncertain and has moved, so they read many labour market indicators rather than one number. But they argue the mechanism is real: when demand outruns the economy's capacity, businesses compete for workers, and wages and prices accelerate. Critics of the Job Guarantee add that its extra income is still spent in the wider economy and that a natural rate of unemployment would still apply below which inflation rises.

Where is the actual disagreement?

Mostly about mechanism and evidence: how stable the link between unemployment and inflation is, and whether a pool of people employed at a fixed wage would restrain inflation as well as a pool of unemployed people. There is also a values question about how to weigh the costs of unemployment against the costs of inflation.

What evidence matters?

Question for you

Explain in your own words why the RBA thinks some unemployment may be unavoidable when fighting inflation, and what MMT economists propose instead.

Your answer (stays in this browser)

Show a model answer

The RBA thinks that when unemployment falls below a certain level, businesses compete for scarce workers, wages rise faster and inflation builds, so bringing inflation down usually means some rise in unemployment. It admits that level, the NAIRU, is uncertain. MMT economists argue that the economy's buffer against inflation should be people employed in a Job Guarantee at a fixed wage, not people without work. The disagreement is whether that buffer would really hold inflation down, and how to weigh the costs of each.

Saved in this browser only. It records that you marked it, nothing more.

Where this leads. If a currency-issuing government could afford a Job Guarantee, what are taxes for? The Tax topic takes that up.

Go further (optional, more advanced)

Estimating the NAIRU and the Unemployment Gap, Tom Cusbert, RBA Bulletin (June 2017) (opens in a new tab)
In Depth: Full Employment, RBA Statement on Monetary Policy (February 2024) (opens in a new tab) How the RBA actually estimates the NAIRU, and how uncertain the estimate is.
Working Future: The Australian Government's White Paper on Jobs and Opportunities, Treasury (2023) (opens in a new tab) The government's current full employment objective and strategy.
Australia's new White Paper on Full Employment is a dud and just reinforces the failed NAIRU cult, Bill Mitchell (2023) (opens in a new tab) Mitchell's critique of the 2023 White Paper, for the MMT side of the same document.
Public Service Employment: A Path to Full Employment, Levy Economics Institute (2018) (opens in a new tab) The detailed US modelling Tcherneva cites, with cost, output and inflation estimates.
Would a job guarantee be work for the dole 2.0?, David Sligar and Hugh Sturgess, Inside Story (2020) (opens in a new tab) The strongest Australian critique, written for a general reader.

Sources for the key ideas