A PROGRESSIVE AGENDA FOR TACKLING AUSTRALIA’S PRODUCTIVITY CRISIS
Inside Story, 29 July 2019
At the start of June, the Productivity Commission quietly dropped a bombshell. Australia’s productivity growth had basically stalled. Labour productivity — output per hour worked — was more or less flatlining. After a generation in which labour productivity had grown at almost 2 per cent a year, it had tumbled to just 0.2 per cent.
The commission called the results “mediocre” and “troubling,” but for some sectors they were downright appalling. In farming, mining, construction, transport and retail, labour productivity went backwards. In other words, workers in those sectors were producing less per hour than they had the year before. The latest numbers continued a trend of weakening productivity growth that the commission dates back to 2013.
To understand why Australia’s productivity crisis is so serious, it’s worth recognising why productivity matters. Through Australia’s history, our economy has become massively more productive. Australian workers today produce nearly four times as much output every hour than in the 1960s. This has been a central driver of rising living standards.
Productivity measures how efficiently the economy turns labour and capital into goods and services. When the Australian economy becomes more productive, we are producing more output from a given level of inputs. Higher productivity creates the potential for household incomes to rise faster than the rate of inflation. A more productive economy can be more generous to the disadvantaged, can reduce its impact on the natural environment, and can play a bigger role in international affairs.
Productivity doesn’t automatically bring fairness: in recent times, workers haven’t received their fair share of the modest productivity growth delivered by the economy. But without rising productivity, wages will eventually stagnate and living standards will stop increasing. Whether your priority is longer lifespans or lower taxes, raising Newstart or building motorways, you should be in favour of productivity growth. Productivity is the engine of the economy, and right now, that engine is making a nasty rattling noise.
A few weeks after the Productivity Commission delivered its damning annual update, a group of boffins gathered together in Sydney for a conference about productivity. Convened by the OECD, a global think tank for advanced countries, the event brought together international experts to discuss the problem and suggest solutions.
For Australia, the most hard-hitting presentation came from Treasury’s Meghan Quinn, who revealed that researchers in her department, led by Dan Andrews, had been investing in a new analysis that links together workers and firms, and delving deeply into fresh data about the dynamics of the Australian economy. Since 2002, Quinn showed, the most productive Australian firms (the top 5 per cent) had not kept pace with the most productive firms globally. In fact, Australia’s “productivity frontier” has slipped back by about one-third. The best of “Made in Australia” hasn’t kept pace with the best of “Made in Germany,” “Made in the Netherlands” or even “Made in America.”
And then there’s the other 95 per cent. For these firms, productivity might as well be a movement in favour of giving ducks more activity. In the past two decades, their output per hour worked has barely risen. In other words, nineteen out of twenty Australian firms don’t produce much more per hour than they did when Sydney hosted the Olympics,
What’s going wrong? Part of the problem is that many firms aren’t investing in new technologies. Less than half have invested in data analytics or intelligent software systems. Only three in five have invested in cyber security, making them vulnerable to hacking and ransomware attacks.
It’s not just that companies aren’t investing simply in technology — they’re not investing in anything at all. This year, the Productivity Commission had to use a new term in its report. Typically, the commission measures how the amount of capital per worker has increased — a concept known as “capital deepening.” This year, for the first time in record, the amount of capital per worker went backwards. The economy had experienced “capital shallowing.” Given that capital deepening has accounted for about three-quarters of labour productivity growth, this is frightening.
Across the economy, businesses are cutting back on research and development and investing less in good management. The share of firms that are “innovative” is no longer growing. A survey of management practices in manufacturing firms found that Australia’s managers rank below those in Canada, Sweden, Japan, Germany and the United States.
It’s been said that you could already tell in the 1950s that Detroit would one day suffer a crash. Although automakers were thriving, the city lacked startups. Once the traditional car manufacturing plants got into trouble, the city slumped. What is true for Detroit holds for cities, regions and countries across the globe: newborn firms are as critical to an economy as newborn babies are to a society’s demography, bringing fresh approaches, shaking up existing industries, and offering new opportunities to workers.
Yet for all the talk of Australia as a “start-up nation,” our new business creation rate isn’t accelerating. In fact, our start-up rate seems to be stopping, though it’s partly masked by a quirk in the way we measure new businesses. The conventional start-up figures include anyone who registers for an Australian Business Number. This means that when a public servant takes a voluntary redundancy, only to come back the next month as a consultant, he is registered as a new business. Likewise when a tradie is “encouraged” by her boss to become a sham contractor. Neither of these cases involve true business formation, so they distort the data.
The way to get around this issue is to look only at “employing businesses’: firms that hire at least one worker. On this metric, Treasury estimates that the new business formation rate in the early 2000s was 14 per cent a year. Now, it’s down to 11 per cent a year. Strip out non-employing businesses and it turns out that our economy simply isn’t hatching new firms like it used to.
Another sign that the economy may be stagnating comes from figures on job-switching. A common myth is that changing jobs is bad for workers, and is happening more frequently. In both cases, the reverse is true. Workers who switch jobs typically experience a significant pay increase. In fact, if you study wages over a career, the largest salary rises tend to come when employees switch firms. Occasionally, job changes will be involuntary and painful — but more often they are voluntary and beneficial.
To see why, imagine for a moment that Australia instituted a rule saying that no one can switch jobs. Anyone who doesn’t like their boss or wants to try working in a different sector wouldn’t be allowed to make the change. Growing companies couldn’t attract workers from their competitors. Such a rule would be profoundly anti-worker. Consistent with this, Treasury’s analysis finds that a drop of one percentage point in the job-switching rate is associated with a 0.5 percentage point drop in wage growth across the economy.
While changing jobs tends to benefit workers, it is happening less often than in past decades. Forget what you’ve read about a fast-churning labour market and the end of “jobs for life”; workers are stay longer in their jobs. In the early 2000s, the rate of job switching was 11 per cent a year. Now, it’s down to 8 per cent. It’s not the fault of employees: there are simply fewer good opportunities available. According to Treasury’s analysis, much of the drop in job-switching is because workers are less likely to transition from mature firms to young firms. With fewer start-ups firms, it stands to reason that there are fewer start-up jobs.
What is to be done? Some people see productivity as a matter of cutting: cutting protections for employees, cutting environmental regulations and cutting the social safety net. Yet when the Productivity Commission was commissioned by the Coalition to compile a list of reforms that might raise national productivity, weakening workplace protections didn’t appear among its top twenty-eight recommendations. Rejecting claims by groups such as the Institute of Public Affairs and the Minerals Council of Australia, the commission’s 2017 Shifting the Dial report noted that “most of the workplace relations law works well to get the balance right between the desires of firms for a fully flexible resource and the need to protect workers from exploitation.”
By contrast, a progressive plan to raise productivity would recognise that the productivity slump has coincided with the marked rise in inequality over the past generation. Earnings inequality, household income inequality, wealth inequality and top income shares have all risen in Australia since the 1970s. In several advanced countries, including Australia, productivity growth has outpaced real wage growth: a problem the OECD calls “the productivity–inclusiveness nexus.” Ensuring that workers get a fair share of productivity gains isn’t just a matter of fairness. Middle-class growth supports consumer demand, which in turn allows businesses to continue to grow. It is not a coincidence that household consumption is languishing at the same time as real wage growth has slowed.
A progressive agenda for raising productivity would fall into three categories: investing in individuals, investing in infrastructure, and investing in institutions.
Investing in individuals: Education is a critical component of productivity. The median woman with a bachelor degree earns roughly $800,000 more in a lifetime than a year 12 graduate who completes no further study. For men, the lifetime difference is $1.1 million. This represents a 65 per cent earnings boost for men and an 80 per cent earnings boost for women.
Graduates are more likely to start new enterprises, and more likely to engage in social entrepreneurship. Education also has positive spill-over effects, with better-skilled employees raising the productivity of their co-workers.
Over the past generation, the average educational attainment of the workforce has significantly increased, with Australians more likely to complete year 12 and more likely to attend university. But the share of people completing an apprenticeship or traineeship has collapsed, and is currently at its lowest rate since at least 2005. Scandals among private vocational education providers have eroded confidence in that system.
At a school level, completion rates are up but we have a massive challenge with test scores. The OECD’s PISA tests show a troubling trend, with teenagers’ scores in literacy, numeracy and science declining significantly since the turn of the century. This continues a pattern that Melbourne University’s Chris Ryan and I documented some years ago, when we showed that teens’ literacy and numeracy levels had failed to rise between the 1960s and the 2000s.
At a university level, arbitrary caps on domestic student places were removed several years ago. The principle was simple: if a young person is talented enough to complete a degree, why not allow universities to make a place available? The policy significantly expanded university places, and therefore the productive potential of the economy. But since the beginning of 2018, the federal government has frozen university grants, which has effectively ended the demand-driven system. This has particularly hurt universities with expanding enrolments, and has hit students who want to study in disciplines that rely heavily on government funding, particularly engineering, sciences and allied health.
In a changing labour market, we should aspire to an Australia in which all young people get a great high school education and a post-school qualification. A much stronger focus on teacher quality would improve the performance of Australia’s schools. Pre-apprenticeships can reduce dropout rates in vocational education (currently half of those who start an apprenticeship do not finish). Restoring the demand-driven funding system would enable almost 200,000 more Australians to attend university by 2030. Many of those new students would be Indigenous, from low-income families, or the first in their family to attend university.
To be productive, people also need to be healthy. As the Productivity Commission’s Shifting the Dial report highlighted, it is also vital to ensure that the healthcare system is run as productively as possible. Medicare needs to adapt to provide comprehensive care that rewards general practitioners and other providers for managing complex cases of chronic illness. An examination of the primary care system by the Grattan Institute found that the dominant Medicare fee-for-service model encourages “reactive rather than systematic care,” and that “much greater emphasis needs to be placed on service coordination and integration for people with chronic disease.” Where interventions have been identified as low value, more effort should be put into informing healthcare providers, including through “do not do” lists. Electronic health records can also improve the productivity of the healthcare system by avoiding duplicate tests and ensuring that physicians have the information they need to make the right decisions.
Investing in infrastructure: As technologies advance, no piece of infrastructure is more important than high speed internet. The difference between fibre-to-the-premises and fibre-to-the-node is critical for applications such as cloud computing, high-definition videoconferencing, telemedicine and distance education.
Choosing to replace the fibre rollout with a multi-technology mix has led to services that are slower, less reliable and more expensive. Time without internet connectivity is a critical source of lost productivity for students and professionals working from home. Unreliable copper services are letting many companies down. The immediate needs for the NBN sound like the advice parents might give a wayward teen: more fibre, higher standards, and proper accountability when providers fall short. It’s also vital that Australia gets the rollout of 5G right: providing the enabling infrastructure not only for smartphones, but also for the Internet of Things.
Open data innovation can also be a major driver of economic growth. Four-fifths of American smartphone users use an app that relies on open data every day. The British Open Data Institute has identified open data–driven businesses that employ more than 500,000 people and turn over £92 million. New Zealand’s Integrated Data Infrastructure approach, for example, combines anonymised information on health, education and crime to offer insights to policymakers and researchers. Governments at all levels should be identifying high-value data sets that can be de-identified and made available to boost productivity.
Governments also need to improve their data policies. Reinventure’s Danny Gilligan points out that government policies on privacy and cybersecurity are like a brake, while innovation policies are like an accelerator. Yet unlike a car, governments often put the brake and the accelerator a long way away from each other. “Brakes” like the Critical Infrastructure Centre don’t spend enough time talking with “accelerators” like Data61. Gilligan contrasts the situation with Singapore, which coordinates how government engages with data economy policies to minimise the costs and maximise the benefits. Data could be a significant source of productivity gains for the Australian economy in coming decades, but only if we get the infrastructure right.
Information superhighways aren’t the only kind of highway. Traffic congestion erodes our quality of life and acts as a handbrake on productivity. A fast-growing nation clearly needs better road networks, urban public transport projects and additional parking spaces at public transport hubs. But the answer isn’t merely to “build more stuff,” it’s to build the right stuff. Infrastructure spending must be based on economic cost-benefit analysis, not political calculus. This means giving greater focus to projects that are on the Infrastructure Australia priority list, and for which a business case has been completed.
Real social gains can sometimes come from arrangements such as value capture and public–private partnerships, but the trickiest infrastructure question is “should we build it?,” not “how do we finance it?” If the benefits don’t exceed the costs, transport infrastructure is a bad idea, regardless of whether it will be paid for by today’s taxpayers or tomorrow’s taxpayers. And there’s a semitrailer barrelling towards us: as the Productivity Commission’s Shifting the Dial report notes, road user charging is set for a shakeup whether we like it or not. Right now, the system depends almost entirely on fuel taxes, which will disappear with the advent of electric cars.
Although a smooth transition to clean energy is critical to maintaining strong productivity growth, Australia’s emissions and energy prices are rising. That contrasts with twenty-one other nations — including the United States, Britain, France and Germany — which have decoupled their carbon pollution from economic growth since the start of the century.
Inevitably, our energy system will move from old coal generation to gas generation and renewables. A more productive energy system will improve the productivity of the whole economy. Delaying the transition will only push up prices and increase pollution. As the Reserve Bank warns, climate-related losses pose a risk for businesses and households, and financial stability “will be better served by an orderly transition rather than an abrupt one.”
Straightforward energy reforms can be introduced. The Productivity Commission estimates that reforms associated with the electricity transmission network — such as critical peak pricing and the rollout of smart meters — could generate large efficiency gains. It points out that plenty of other nations have replaced piecemeal solutions with a single price on carbon. It recommends that governments more clearly articulate the trade-off between reliability and cost. And it suggests that we get pricing right, so that producers pay for additional costs they impose on the system (such as frequency management), and users pay for access to the grid (so that people cannot simply use it as a back-up system).
Good institutions: The third part of the progressive productivity agenda is to improve institutions so that they support a more productive economy. This starts with how government institutions support innovation. As past Australian Innovation System reports have noted, innovation and adaptation to technology are vital for productivity growth. However, the 2018 Global Innovation Index ranks Australia just twentieth in the world. Since the global financial crisis, the volume of venture capital investment has fallen by nearly two-thirds. Innovation collaboration between government, business and academia is less common in Australia than in many other OECD nations.
The economic rationale for subsidising research and innovation is that the community make gains beyond the private benefit to the developer. Federal government funding for science, research and innovation through grants and tax subsidies amounts to at least $10 billion annually. But right now, there isn’t much incentive for firms to work with universities. The government has been slow to implement its proposed Consumer Data Right. Educational bodies and disability groups have complained about the stultifying effect of a copyright law that lacks fair use exemptions. Ironically, innovation policy isn’t very innovative, since the federal government doesn’t devote enough energy to evaluating the impact of its many innovation policies.
The institutions that support trade and investment are also essential to improving productivity. As a medium-sized economy, Australia’s productivity performance is invariably intertwined with our engagement in the region. World trade is just another form of comparative advantage, letting countries specialise in what they do best. Just as your hairdresser doesn’t defeat you when you get a haircut, Japan doesn’t defeat you when you buy a PlayStation. Sellers aren’t vanquishing buyers — both are benefiting from specialisation.
We must do openness better. When it comes to trade, it’s vital to recognise that the best type of trade agreements are multilateral agreements, followed by regional and then bilateral. Bilateral deals can have benefits, but can also distort trade. One way of ensuring that trade agreements are in Australia’s economic interest would be to allow the Productivity Commission to scrutinise them before signing, and again a decade after they come into force. This would provide some certainty that vested groups had not captured the negotiation process, and ensure that if we make mistakes, we learn from them.
On foreign investment policy, it would be worth reviewing the plethora of screening thresholds. It is difficult to mount an economic justification for requiring the Foreign Investment Review Board to approve a $300 million business acquisition by a Canadian investor but not a US investor. It would help the public conversation on foreign investment if the Treasury set out reasons for all significant foreign investment decisions: both acceptances and rejections.
We can also be more productive through an institutional push to improve Asian engagement. As CEDA notes, Australia has less outbound foreign direct investment in China, Japan, Korea, India and all ASEAN countries combined than it does in New Zealand. Among year 12 students, only one in fifty study Chinese. Fewer year 12 students study Indonesian than in 1972. AsiaLink found that more than half of all ASX200 board members demonstrated little or no knowledge of Asian markets. If we are to grow services exports to our region, a higher level of Asia literacy is essential: from the classrooms to the boardroom.
And then there’s the institutions that govern markets. In uncompetitive markets, firms have a weaker incentive to pursue productivity gains. One British study found that a 25 per cent increase in market concentration leads to a 1 per cent fall in productivity. Another study attributed about one-fifth of productivity growth to better competition policy.
Over recent decades, several sectors have become significantly more concentrated. The annual volume of mergers has increased fivefold since 1990. At the same time, as we’ve seen, the new business formation rate has fallen. The result is a significant increase in market concentration across key industries in Australia. One rule of thumb is that a market is excessively concentrated if the largest four firms control more than a third of it. Under this definition, ANU’s Adam Triggs and I found that over half of Australian industries are overly concentrated. In department stores, newspapers, banking, health insurance, supermarkets, domestic airlines, internet service providers, baby food and beer, the biggest four firms comprise more than 80 per cent of the market. The World Economic Forum’s Global Competitiveness Report found that Australia does badly on “the extent of market dominance,” ranking us fifty-third in the world.
Compared with many other countries, penalties for anti-competitive conduct in Australia are too low, and our competition watchdog is underfunded. Unlike other countries, the Australian Competition and Consumer Commission lacks a market studies power, meaning it can’t use investigatory powers to explore public interest issues such as pricing discrepancies and increased market concentration. And after approving a merger, the commission has no systematic process of deciding whether or not it made a mistake. Like a coach who watches the video replays, post-merger reviews of productivity, wages and prices could help improve decisions in the future.
Tax institutions matter too. Good tax reform involves closing loopholes. As the late Harvard economist Martin Feldstein liked to point out, winding back tax concessions raises revenue more efficiently than increasing tax rates. In economic jargon, closing loopholes has a lower deadweight cost than raising rates. Yet, as the 2019 federal election showed, the economics are easier than the politics. Every loophole in the personal and corporate tax system has its ferocious defenders. If Australia wants to increase productivity, we need to consider whether we can do so with a tax system that has more holes than a block of Swiss cheese.
Another key set of institutions are those governing management quality. Firms with a healthy management culture are places where employees look forward to arriving at work, where people respect one another, and where diversity flourishes. The best companies listen to their employees, implement good ideas regardless of their origins, and aren’t afraid of change.
The quality of management directly shapes the ability of businesses to adapt and innovate. Managers who lack appropriate technical and personal skills can’t provide the leadership that their firm needs to find continual productivity improvements. Managers who face overly short-term incentives may fail to focus on sustainability and longer-term productivity gains. Discrimination and unconscious bias may lead organisations to overlook talented applicants for appointment and advancement.
There are many creative ways companies can draw on the ideas of their workforce. In one large New Zealand dairy company, for instance, the union initiated a management improvement system with the aim of boosting productivity. Workers were trained in productivity measurement, and the concepts underpinning productivity growth, such as change management and improved teamwork. From their first day on the job, they are encouraged to take responsibility for raising quality, reducing waste, and even considering whether new products can be made with material that is currently being discarded. Managers play a role more akin to coaches than commanders, urging workers to think about improving the way the firm is run. Over a five-year period, the company estimates that the system has produced savings of $100 million, largely through reduced downtime and wastage. The firm has raised its output and quality, but without any loss of jobs.
On a systemic level, a number of German companies foster productivity growth by having worker representatives on company boards. There is even a word for it — mitbestimmung — meaning worker participation in a company’s decision-making. Many of these firms find that a less confrontational approach produces significant efficiency gains. This occurs to some extent in Australia through our industry superannuation system. Industry superannuation funds are established jointly through employer and employee representatives, and as they become significant investors, they have the ability to encourage corporate managers to make better long-term decisions.
There is some evidence that employees in worker-owned firms, such as cooperatives, are more productive and more satisfied. Government policy could do more to foster the growth of cooperatives and mutuals, by facilitating greater access to capital for such firms, and access to government grants, particularly for Indigenous cooperatives under the Indigenous Advancement Strategy program.
Productivity gains also need to be shared. Just as business owners would have little incentive to invest in productivity-boosting improvements if none of the benefits translated into higher profits, so too workers have less incentive to support measures that increase productivity if they do not lead to higher wages.
In the 1970s, real wage growth outpaced productivity gains. Economists called it “the real wage overhang,” and the solution was to moderate wage growth so that it came back into line with labour productivity. Today, the economy faces the reverse problem. Even the modest increases in labour productivity that the economy has been producing haven’t flowed into workers’ pay packets. Australia is now experiencing a “real wage underhang.”
As in other nations, this decoupling of wage gains from productivity growth has led to a fall in the labour share of national income. Some estimates suggest that the drop in the labour share since the 1970s might have been as large as 15 percentage points.
One of the theories for the drop in the labour share is that monopolies don’t need as many employees. As economist Jan Eeckhout argued at a recent Reserve Bank conference, “market power depresses the demand for labour: firms set higher prices and therefore they produce less output, for which they need less labour. For the labour market to be in equilibrium, the economy moves along the upward sloping labour supply curve until a new, lower wage is obtained.” Other research at the same conference found that the wage slowdown was especially pronounced in sectors that don’t face international competition. Economists have long known that monopolies hurt consumers — now it appears they can harm workers too.
Left unchecked, this trend threatens to undermine the incentive for employees to continue to pursue productivity gains. A more collaborative approach to industrial relations is likely to be in the long-term interests of capital as well as labour. As economist Saul Eslake points out, corporate managers aren’t judged on their share of profits, but on their actual profits. Firms would be better with a smaller share of a growing pie than a larger slice of a shrinking one.
Government institutions often pay lip service to evidence-based policy. Instead, they’re often driven by the idiosyncratic ideas of a few people in charge — what has been called “eminence-based policy.” In my book Randomistas: How Radical Researchers Changed Our World, I argue that agencies should be more modest about theories, and more willing to rigorously evaluate programs. One way of doing this would be to systematically conduct randomised policy trials, using treatment and control groups to test government programs in the way that companies test new pharmaceuticals. Already, randomised trials have provided unexpected insights about how to reduce recidivism by drug-offenders, how to use video feedback in teacher training, and how to encourage new business formation in developing nations. Building a better feedback loop helps drive continuous improvement in the productivity of government.
Too often, Australians see productivity as a dirty word — synonymous with working harder, rather than working smarter. But productivity should lead to a better quality of life, in which people have more choices in the workplace and more opportunities to spend time with friends and family. The path towards higher productivity should also allow us to live in a cleaner environment, and to be more generous to the needy. Tackling major challenges, from gender equity to traffic congestion, is easier in a highly productive economy.
Economists talk about the “rule of seventy-two.” If you want to know how long it takes to double living standards, just divide seventy-two by the growth rate. This means that a 2 per cent growth rate doubles living standards every thirty-six years, or about once a generation. But a 4 per cent growth rate doubles living standards every eighteen years, or twice a generation. That’s why we should invest in individuals, infrastructure and institutions, to lay the groundwork for lasting improvements to Australia’s productivity growth rate.
Australia doesn’t have to choose between fairness and productivity. We should be aiming to achieve inclusive growth by sharing productivity gains across the community. This not only creates greater wellbeing, but also supports ongoing reform. If an elite cabal capture all of the benefits of reform, they should not be surprised if the next round of reforms meets a public backlash. The goal should be to raise productivity to the benefit of all Australians.
Andrew Leigh is the Shadow Assistant Minister for Treasury and the Federal Member for Fenner.
Authorised by Noah Carroll ALP Canberra