Talk to a gathering of parliamentarians, business leaders and others from the UK and Australia, in Melbourne, on Wednesday 13th March 2024

Productivity

Productivity – what we get by way of dollar value of goods and services produced, for every hour of work that we put in – as Paul Krugman famously said (before he became a left-leaning columnist for the New York Times), “isn’t everything, but in the long run it’s nearly everything”. In particular, productivity growth is the only sustainable source of long-run improvements in people’s real incomes, and in their standard of living.

Australia experienced a surge in (labour) productivity growth in the late 1990s and early 2000s, peaking at 2.5% pa over the five years to 2002-03. But since then it has slowed to just 0.8% pa over the five years to 2022-23.

Australia’s experience of declining productivity growth isn’t unique – it’s been apparent in most ‘advanced’ economies over the past two decades. So the most plausible explanations for the productivity growth slow-down are most likely factors common to ‘advanced’ economies rather than peculiar to Australia.

Those explanations include an apparent slowdown in the rate at (and even extent to) which advances in productivity at so-called ‘frontier firms’ (which invent new technologies) are diffused to ‘laggard firms’ in the same industry or in different industries (which adapt and use them). This may be in turn partly inherent in the nature of the new technologies being developed at ‘frontier firms’, but almost certainly owes something to the ‘market power’ which ‘frontier firms’, especially in the ICT area, are able to exercise, including by being allowed to ‘swallow up’ potential rivals before they get large enough to mount a serious challenge to the established position of those ‘frontier firms’.

Another important explanation is an apparent decline in the ‘dynamism’ of ‘advanced’ economies – that is, in the movement of labour and capital from less productive to more productive uses, both within and between industries. This is evident in Australia in the decline in rates of business entry and exit, and in the decline in ‘job-switching’ by employees – to which the proliferation of ‘non-compete clauses’ in employment contracts may have contributed – paralleled by an increase in the average length of time which employees remain with the same employer.

Declining competition and increased monopoly power in many markets, in Australia and in other ‘advanced’ economies appears to have been a factor as well. When firms can increase profit margins by increasing mark-ups, they have less incentive to innovate, or to increase productivity.

And then there is the almost complete dearth (in Australia) of any productivity-enhancing reforms, under governments of both major political persuasions, at both the federal and state levels, since the late 1990s – as Paul Keating rightly bewailed on the occasion of his 80^th birthday.

But it’s not simply the absence of productivity-enhancing reforms that has been a problem, in Australia at least. There have also been a series of productivity-detracting policy measures enacted by governments in pursuit of other objectives.

I’m not talking about the plethora of productivity-reducing measures that have been instituted by successive governments with the purported aim of enhancing ‘security’ at airports and in countless other places over the past 22 or so years – much as that’s been a favourite hobby-horse of mine over that interval.

Let me instead give you two instances of what I mean by ‘productivity-detracting policy measures’. First, what I call ‘manufacturing fetishism‘ – the apparent belief that manufacturing is an inherently more noble or worthy form of economic activity than agriculture, mining, or any kind of service provision, and that manufacturing jobs are more important to retain or create than any jobs in any other sector.

This belief has informed a range of bad policy decisions over the past two decades (indeed, it was a major driver of some very bad policy decisions from Federation until the late 1980s).

Gross value added in manufacturing per hour worked in Australia in the 2022-23 financial year was $79.20. That’s $9.88 per hour, or 11.1%, below the average for all industries. In other words, while manufacturing may be a ‘high-productivity’ sector in some other economies, in Australia it’s a low-productivity form of economic activity.

By definition, therefore, policy interventions designed to support more activity or employment in manufacturing than would otherwise be the case detract from overall labour productivity.

As an aside, labour productivity in South Australian manufacturing is $13.80 or 18.2% lower than productivity in manufacturing Australia-wide. So policy interventions designed to promote manufacturing in South Australia – otherwise known as defence procurement policy – have an even bigger detrimental effect on overall labour productivity.

The second type of ‘fetishism’ that I think has adversely impacted Australian productivity growth is what I call ‘small business fetishism’ – the seemingly widespread, almost universal, belief that there is something inherently more noble or worthy about running a small business than about working for a big one, a government agency or a not-for-profit, and that therefore people running small businesses deserve a range of subsidies not available to people in otherwise similar circumstances, and should pay less tax on any given amount of income than people earning the same income in the form of (for example) wages or salaries.

Such preferential treatment is routinely defended on the basis of assertions that “small business is the ‘engine room’ of the economy’.

Excuse me, but that’s (to use an expression that will be familiar to our British friends here today) bollocks.

Yes, I know, small business accounted for 41.7% of total private sector employment in 2021-22. I also know that 10 years earlier, small businesses (those with fewer than 20 employees) accounted for 43.3% of total private sector employment. And that fifteen years earlier it accounted for 49.9% of total private sector employment. Over the past decade, small businesses have accounted for just 20.4% of total private sector employment growth (ie less than half their share of total employment).

I also know that labour productivity in small businesses is 19% lower than the average for all businesses.

Even if you exclude the mining sector, which is dominated by large businesses and in which labour productivity is exceptionally high (because it’s so capital-intensive), labour productivity in small businesses is 12% below the average for all businesses.

So, simply as a matter of arithmetic, policies which favour the retention of labour and capital in small businesses simply because they are small and for no other reason – such as the 5 pc point discount on the company tax rate payable by companies with annual turnover of less than $50 million, preferential tax write-offs for investments by small businesses, “entrepreneur’s tax offsets”, exemptions from payroll tax for small businesses and so on, not only do nothing to create jobs or spur innovation, but actually detract from productivity growth.

Preferences for small businesses, simply because they are small, should be replaced by preferences for new  businesses – which of course will almost always be small, but which are also much more likely to create jobs, much more likely to innovate, and which cannot help but eventually become ‘old’ businesses and hence no longer eligible for preferential treatment, unlike many small businesses which choose to remain ‘small’ so as to remain eligible for preferential tax treatment and access to a range of subsidies.

Two other points to note about recent trends in productivity in Australia.

First, there has been a significant shift in the composition of employment towards intrinsically low-productivity industries  – such as health care and social assistance, retail trade, accommodation and food services, and art and recreation services. The share of total hours worked which were worked in the seven sectors where the level of labour productivity is more than one-third below the all-industries average rose by 3 pc points between 2012-13 and 2022-23. This is in line with consumer and community preferences, so it would be wrong to seek to over-ride it: but it underscores the need to foster faster growth in labour productivity in other sectors if it is not to result in a permanent reduction in the rate of overall labour productivity growth.

Second, there’s something ‘odd’ happening to labour productivity in the mining sector, which has dropped by 23% since the December quarter 2019, reflecting an increase of 19% in mining industry employment (and a 23% increase in total hours worked in the mining sector) but a 3.6% drop in mining industry output. I’m not sure what’s going on there, but it has had a significant impact on overall labour productivity – subtracting more than 3% from the economy-wide level of labour productivity over the past three years.

Tax reform

There haven’t been any successful serious attempts at tax reform since the Howard-Costello reforms of 2000, which saw the replacement of the wholesale sales tax, and a range of inefficient and (as the High Court had ruled) unconstitutional state taxes replaced with the GST, accompanied by wide-ranging reductions in personal income tax.

There have been some serious attempts at tax reform which weren’t successful, such as Joe Hockey’s aborted ‘Re:think’ exercise, and some changes to individual taxes which were ‘successful’ in the sense that they were implemented, but which (in my opinion) don’t amount to ‘tax reform’.

I include in that category the repeated rounds of personal income tax cuts in the last two terms of the Howard Government, the substantial increase in the tax-free threshold under the Gillard Government, the introduction and abolition of the carbon tax, the 5 pc point reduction in the company tax rate for “small” businesses introduced by the Turnbull Government; and the ‘Stage 1, 2 and 3’ tax cuts introduced by the Morrison Government, and the re-casting of the last of those by the Albanese Government.

The Howard-Costello tax reforms of 2000 were ‘successful’, politically as well as economically, because the Government was able to say, truthfully, that all Australians were better off as a result of them.  And that was possible because the Howard Government was able to draw down on a very large Budget surplus in order to ensure that no-one was worse off, and indeed everyone was better off, either through reductions in income tax or increases in social security benefits which more than compensated for the impact of the GST on their disposable income.

That isn’t possible today, and seems unlikely to be possible at any time in the foreseeable future, unless you believe that the iron ore price is going to remain above US$100/tonne forever (more on that later).

Rather, it seems to me that any sustainable reform of Australia’s taxation system is going to have to entail making at least some people worse off, in order to raise additional revenue to fund the additional spending that the Australian people clearly want, or which they’re going to get, whether they want it or not.

What I mean by that is that it seems very clear, from all the medium- and long-term projections of budgetary aggregates that have been published by governments of both political persuasions since 2020 is that Commonwealth Budget payments are going to constitute around 26½% of GDP for the foreseeable future – which is about 1½-1¾ pc points of GDP more than the average between 1975-76 and 2019-20.

That’s partly because the Australian public clearly wants more spending on health, aged and disability care – and woe betide any political party aspiring to form government which suggests it might not deliver it – and, additionally, whether they want it or not, the Australian public is going to get more spending on defence, and on public debt interest.

I can’t think of any politically saleable way of sustainably reducing Commonwealth Budget payments in other areas by the equivalent of 1½-1¾ percentage points of GDP.

More importantly, nor could the Abbott, Turnbull or Morrison Governments during their nine years in office.

So unless we are willing to fund that additional government spending by borrowing – that is, by running deficits of the order of $50 bn a year in the mid-2020s rising to almost $70 bn a year by the mid-2030s – any serious tax reform exercise has to include some way or ways of raising an additional 1½-1¾ pc pt of GDP in revenue.

It’s a telling indication of how dysfunctional our tax system has become, that personal income tax is now taking a record proportion of personal income, and of GDP, and so is company tax: yet total Commonwealth taxation revenue is still less, as a proportion of GDP, than the record high attained in 2000-01.

We can’t fill the hole on the revenue side of the Budget by lifting rates of personal or company income tax – at least, we can’t fill it that way without a raft of adverse and unwanted consequences. Rather, we should be looking at broadening the base of our tax system, including income tax but also other taxes.

Our top rate of personal income tax is in the middle of the range of ‘advanced’ economies. But the income level at which the top marginal rate becomes payable is very low by OECD standards. For example the top tax threshold for a single taxpayer in the US is about A$780,000 at current exchange rates; in the UK it is about A$286,000; and in Germany it is about $444,000.

But one of the striking things about Australia’s income tax system is that a lot of income which in theory should be taxed at the top rate, in practice isn’t. In 2020-21, according to the ATO’s Taxation Statistics, 47% of the taxable income accruing to people in the top tax bracket (as opposed to just 17% of people whose taxable income is less than $180,000) is in some form other than wages and salaries – that is, capital gains, dividends, rental income, business income, income from superannuation funds, and income received through trusts    – which is typically taxable at less than the marginal rate applicable to wage and salary income.

So a desirable tax reform – one which would be economically efficient (in the sense of reducing artificial incentives to convert income into forms which attract lower rates of tax), equitable (in the sense of treating more equally people in similar circumstances), and politically ‘saleable’, would be to raise the top tax threshold to something more in line with that in other comparable countries, but to also ensure that all income above that threshold was taxed at the top rate.

That might even raise enough revenue to fund a reduction in the top tax rate to something which, in Paul Keating’s words, was less “confiscatory”.

Another sensible reform of the income tax system would be to reduce, or even abolish, Australia’s unusually high tax-free threshold.

There are obviously sound reasons of equity as to why people on very low incomes should not be expected to pay income tax. But there are better ways of achieving that outcome than exempting the first $18,200 of everyone’s taxable income from tax – for example through something similar to the US’ Earned Income Tax Credit, or the Low Income Tax Offset.

Why should the first $18,200 of someone earning (say) $250,000 be completely free of tax – especially if by taxing that first $18,200 (which would have absolutely no adverse incentive effects) it might then be possible to reduce the tax rate on the last (say) $50,000?

Such a reform would necessarily entail reducing the capital gains tax discount (which is way too generous), getting rid of negative gearing, reducing the generosity of the tax treatment of superannuation fund earnings and/or payments out of superannuation savings (which again are more generous than they need to be to encourage people to save for their own retirement), and taxing trusts as companies.

In principle, I’d be in favour of broadening the base and/or raising the rate of the GST. The revenue raised by Australia’s GST represents only about 3¼% of GDP, or a bit of 11% of total tax revenues, compared with the OECD average of about 6½% of GDP and more than 20% of total tax revenues.

But increasing the rate or broadening the base of the GST isn’t going to fly if the Federal Government is going to have to wear the odium of raising it, and carry the cost of ‘compensating’ pensioners and other low-income earners for the impact of raising it on them (something which hasn’t been deemed necessary in other countries), only for the states and territories to get the kudos for spending the resulting revenues.

And I certainly wouldn’t support raising the GST or broadening its base, only for an unjustified share of the resulting revenue to be handed over to the only government in Australia (indeed one of very few in the world) which is running persistent budget surpluses, so that it can run even bigger ones, and so that the citizens of the richest state in Australia can enjoy better public services and lower state taxes than other Australians – which has been the outcome of what I regard as the worst public policy decision of the 21st century thus far, namely, the corruption of the long-standing arrangements for distributing the revenue from the GST among the states and territories by the Morrison Government (with the support of the then Labor Opposition) in order to retain (or gain) House of Representatives seats in Western Australia, at a moment in time when they were critical to the outcome of two federal elections.

This decision undermined the purpose of the biggest single spending program in the federal Budget – $92bn in 2023-24, cf. $58bn for the second-biggest program, ‘support for seniors’, and $42bn for the third-biggest, the NDIS.

That intent, first promulgated in the aftermath of the 1933 ‘Wexit’ referendum, was to ensure that, as far as possible, each state and territory had the capacity to provide a similar range and standard of public services to its citizens whilst levying on them a similar burden of state taxes.

That objective – which is one of the things that has made Australia a ‘better’ country to live in than the United States, or Canada – delivered Western Australia more than (to put in the language that Western Australia uses) ‘more than 100 cents in the dollar’ of whatever ‘untied’ federal grants to the states and territories were going around, for the best part of seven decares.

But when, through no great effort on the part of any Western Australian state government, WA became the richest state in the country (as measured by per capita gross state product), by a much larger margin than any other state had ever been, and hence was required to put into the pot from which it had happily drawn for almost 70 years, threw a giant tantrum and wanted the rules changed to what is, effectively, a system under which “heads WA wins, tails the federal government loses”.

The ‘no-worse-off transitional guarantee’, necessary in order to cajole the other states and territories into acceding to something that was so obviously otherwise against their interests, was supposed to cost the federal budget $8.2bn over eight years, on the assumption that the iron ore price would remain at US$55/tonne over that period. With the iron ore price instead having exceeded US$100/tonne for all but four months since then, and with the Albanese Government having extended the ‘NoWO’ until 2029-30, the cost to the federal Budget has now officially blown out to $39.8bn over 11 years  – assuming the iron ore price drops to US$60/tonne by September. If the iron ore price stays where it is until 2029-30, the cost will be at least $50bn over 11 years. That’s the largest cost ‘blow-out’ of any single policy initiative ever, with the possible exception of the NDIS.

It’s certainly a much larger cost-blow out than the much more well-publicized cost over-runs in ten major defence procurement projects. How many more Hunter Class frigates could we have afforded if we weren’t gifting $40-$50bn over 11 years to Western Australia?

Turning to other taxes, for years ‘every galah in the pet shop’, to deploy another Keatingism, has been calling on state governments to replace stamp duty on land transfers with a more broadly-based land tax. Former NSW Premier and Treasurer Dominic Perrotet – the only state leader to have seriously entertained this proposal – described it as “the most important state economic reform of the last half century” – although in the end, he too put it in the “too hard” basket.

But if the states won’t do it, why shouldn’t the Federal Government impose a federal land tax – as previous Federal governments did between 1910 and 1952 (when it was abolished by the Menzies Government as a sop to the states for the Federal Government having assumed a monopoly of income tax as a ‘temporary war time measure’ in 1942, something which the Menzies Government (like all its successors) was reluctant to reverse).

One final thought on tax reform is, why not re-introduce some form of inheritance tax? Australia is one of only 12 OECD countries that don’t levy inheritance or estate taxes – and among those who don’t are Sweden and Austria, who don’t tax assets on death because they tax assets annually while taxpayers are alive. Both the US and the UK – whose tax systems are the ones most commonly used as benchmarks for Australia’s – do levy estate duties, and have never seriously considered abolishing them. Neither Margaret Thatcher nor Ronald Reagan (and nor for that matter Sir Robert Menzies) ever proposed getting rid of estate duties.

As the OECD (the shop run by former Australian Finance Minister Matthias Cormann) says, “well-designed inheritance taxes can raise revenue and enhance equity, at lower efficiency and administrative costs than other alternatives”. 

Caring for aging baby-boomers in their final years is going to cost a lot of taxpayers’ money over the next thirty years – most of it raised from younger taxpayers who will also be supporting their own children. Why shouldn’t some of that be ‘clawed back’ from the estates of those same baby-boomers who have profited mightily from the same house price inflation which has made it so much harder for succeeding generations to become home-owners?

None of the suggestions I’ve made here are going to be very palatable – especially, I suspect, to this audience.

But the point I make in concluding this discussion is, if you’re not prepared to contemplate measures like these, the only remaining alternatives are some combination of never-ending ‘bracket creep’ so that Australia becomes even more reliant on personal income tax than we already are, or on-going budget deficits which burden future generations with ever-increasing levels of debt. Is that what you really want?

‘Peak China’

For most ‘advanced’ economies, the emergence of China as the world’s second largest economy has been a net negative. Most ‘advanced’ economies (apart from Australia, Norway and, to a degree, Canada) are exporters of manufactured goods and importers of commodities. So the emergence of China as the world’s largest exporter of manufactured goods, and the world’s largest importer of commodities, has put downward pressure on the prices of the things that most other ‘advanced’ economies export, as well as displacing many of the things that they produce for themselves; whilst simultaneously putting upward pressure on the prices of the things that they import.

But Australia is a net importer of manufactured goods and an exporter of commodities.

So China’s rapid growth and industrialization over the past 45 years has put upward pressure on the prices of the things that we export, and downward pressure on the prices of the things that we import.

No ‘advanced’ economy has derived more benefits from the rapid growth and industrialization of China over the past 45 years than Australia. We are one of only six ‘advanced’ economies which runs a bilateral trade surplus with China – and ours is the second-largest such surplus, after Taiwan’s.

Australia’s real gross domestic income is now 50% higher than it would have been had our terms of trade (the ratio of the prices we get for our exports to the prices we pay for our imports) remained where they were in mid-1986, when Paul Keating notoriously said that if they kept heading in the direction they had been, Australia would end up as a ‘banana republic’. China has been, overwhelmingly, the reason why they didn’t, but instead rose by more than 125% since then.

And on top of that, there’s been a 20% boost from the increase in the volume of resources exports since then, most of which has also been driven by China.

There has of course been a ‘downside’ to this. Australia is now more dependent on China, both as a market for our exports and as a source of our imports, than we have been on any single country since the early 1950s (when most of our trade, in both directions, was with what we used to call ‘the mother country’, the UK).

China couldn’t have sought to coerce the US, or the UK, in the way that they sought (unsuccessfully, as it turned out), to bend Australia to its will through the use of trade sanctions between 2020 and 2023 – because they don’t import much from either of those countries.

China imported only US$20bn of goods from the UK last year, compared with US$155bn from Australia: Emperor Qianlong’s 1793 letter to King George III , in which he said his country “has no use for your country’s manufactures”, remains true to this day.

China’s era of unprecedented sustained rapid economic growth is almost certainly over. In 2022, China’s economy grew at a slower rate than that of the world as a whole, for the first time in at least 42 years. And in 2023, China’s economy shrank in US dollar terms, for the first time since 1994.

The slowdown which China is now experiencing has a cyclical dimension to it – in particular as a result of the policy-induced contraction in its real estate sector, which had been a major driver of economic growth between 2008 and 2018. In contrast to every other major economy in both the ‘developed’ and ‘developing’ world, China is experiencing deflation, rather than inflation. Yet the ‘Chinese authorities’, as they’re always referred to, are doing remarkably little, either to ward off deflation or to rekindle economic growth.

But from a longer-term perspective the slowdown in the Chinese economy is primarily structural, rather than cyclical.

Looked at through the “three Ps” (population, participation and productivity) framework that the Australian Treasury has used to derive long-range economic projections in the succession of Intergenerational Reports over the last two decades:

• China’s population has been shrinking since 2021, and will decline at an accelerating pace over the next 50 years (0.7% over this decade, 2.7% over the 2030s, 4.7% over the 2040s, 8.2% over the 2050s, according to UN projections);
• because China’s population is also ageing, the decline in China’s working age population will be even more dramatic (1.6% over the course of this decade, 10.9% in the 2030s and 11.5% in the 2040s);
• as a result, the proportion of China’s population which is employed has already fallen by 5.1 percentage points since it peaked in 2007, and will decline at an accelerating rate over the next three decades;
• China’s labour productivity growth rate has slowed from an average of nearly 9% pa over the two decades to 2010, to between 4½% and 5½% pa over the past four years, depending on whose figures you use (The Conference Board or the International Labour Organization) – reflecting among other things that the process of shifting people out of peasant agriculture into manufacturing (a major source of productivity growth between 1979 and 2015) has now run its course, and the fact that the long decline in the share of the Chinese economy under the control of state-owned enterprises has gone into reverse under Xi Jinping.

Additionally, China’s share of global export markets appears to have peaked – not least because other countries are less willing to allow their domestic markets to be penetrated by Chinese suppliers than they had been when China first entered the WTO. There are to be sure some product markets where China will make further market share gains – most obviously motor vehicles (and in particular electric vehicles) – but these are being offset by declines in market share in other products.

Finally China under Xi Jinping has become more hostile to – and less attractive as a destination for – foreign direct and portfolio investment.  As a result, foreign investment inflows into China have come to an almost complete halt – totalling just $US551mn in 2023, down from a peak of US$521bn as recently as in 2021.

Combined with a shrinking current account surplus (equivalent to just 1.5% of GDP in 2023, down from a peak of 10% of GDP in 2007), the collapse in foreign investment into China raises the risk that China’s fixed exchange rate regime may become untenable should the current account slip into deficit, given the wide divergence between China’s level of foreign exchange reserves (which has been more or less unchanged since the financial crisis of 2015-16) and the continued rapid growth of domestic credit since then.

India is not going to do for the world economy, or for Australia, what China did between the mid-1980s and 2020. It is still too protectionist; it is too suspicious of economies of scale; it wastes too much human capital through various forms of systematic discrimination on the basis of gender, caste and (increasingly nowadays) religion; and it hasn’t invested nearly enough in the basics of health care and education to sustain the sort of growth rates which China did for forty years.

And there are no other Chinas or Indias ‘out there’ – no economies with populations anywhere near as large as China or India, and most of those economies which do seem capable of enjoying rapid economic growth over coming decades (such as Indonesia) are much more self-sufficient in commodities than China and India.

The very likely prospect that China’s economic growth rate will slow to less than 4% pa over the remainder of this decade almost certainly means that commodity prices, and hence Australia’s terms of trade, will decline between now and 2030.

That in turn means that, in the absence of a substantial pick-up in productivity growth, Australia’s real income – and hence the material living standards of Australians – will continue to grow at a slow pace, possibly slower than they did during the 2010s (which was in turn the slowest rate of growth in real incomes over any decade since the 1930s).

The prospect of a substantial decline in commodity prices also underscores how unfair the GST ‘deal’ with WA is – because there is a ‘floor’ under WA’s share of GST revenues (relative to its share of Australia’s total population) when commodity prices are high, but no ‘ceiling’ on its share when commodity prices are low. In other words, ‘heads WA wins, tails the federal government loses’. Another reason why it’s the worst public policy decision of the 21^st century so far, and ought to be reversed.