It’s not news, at least in my circles, that much of the developed world is seeing economic stagnation.
Last month Australia convened a Productivity Summit/Economic Reform Roundtable to try to better address the sluggish growth of the economy relative, in particular, to ever increasing government spending obligations.
Specifically, Total Factor Productivity (TFP) growth measures the rate at which economic output grows faster than capital and labor inputs. Usually attributed to “technology”, but technology must be understood in a very broad way, since it includes things like financial system efficiency and manufacturing management techniques, as well as blueprints and research papers. The boundaries between labor, technology and capital are always a bit fuzzy. The reason TFP is important is that the supply of labor and capital are fundamentally finite, while knowledge is infinite. TFP growth is by far the best mechanism to accelerate the rate at which a society compounds wealth and builds a terrific future.
For reference, here is a chart showing Australian labor productivity growth, showing recent steep declines.
Reporting on the Australian Productivity Summit has suggested that there is little agreement on anything except for the need for increased government revenue. That is, additional taxation. Sorry, tax reform.
Amidst a blizzard of commentary I found not a single mention of the Laffer curve. The Laffer curve is a simple model that describes the relationship between taxation rate and tax revenue. If we begin at 0%, revenue is zero, and increases accrue linearly. But this linear relationship doesn’t hold for long, and at some point revenue peaks. Beyond this point, additional taxation impedes economic efficiency and destroys incentives to work. In the extreme case, we have a command economy which “functions” without price signals and, historically, has always led to stagnation, repression, and both labor and capital flight. Not what we want!
Taxation past the point of peak revenue is particularly pernicious because a) the money is appropriated from definitionally productive parts of the economy, b) its consumption in public service definitionally deprives the private sector of access to that labor and c) excessive government spending distorts the market, damaging price signals and inhibiting the decentralized flow of economic information that is so essential for prosperity.
This is obvious enough in a steady state situation, but we’re not in a steady state situation. In particular, we have an aging population, increasing expectations around public services, inflation, and flatlining productivity in the public sector – a common challenge in sectors without competition and the “creative destruction” it brings.
As shown in the graph below, public sector productivity is at levels first reached in 2001, while the private sector has improved about 50% over the same period. If public sector productivity had kept up, we could achieve the same provision of services we currently enjoy with just ⅔ of the current tax rate. Alternatively, of the $800b Australia raised in tax last year, more than $250b fell into the hole caused by zero productivity increase in 21 years. $10,000 per man, woman, and child in Australia consumed by a lack of public sector dynamism!
Since our goal here is to ensure sustainable high quality public services supported by sustainable and high levels of economic growth, the impact of taxation on growth is extremely important. That is, the goal of taxation should not be merely to maximize government revenue today, but to maximize it over a period of decades, taking into account the necessity for economic growth and, in particular, TFP growth, to ensure that future Australians can look forward to even better publicly funded services.
As shown on the curve above, the point of peak growth is a long way to the left of the point of peak revenue. The reason for this is that by the time revenue has been maximized, the efficiency of government spending and the burden of taxation has already grown to a point where essentially all the excess productive capacity of the economy is spoken for and a nation’s wealth ceases to be able to accumulate. At either extreme of no taxation or full taxation, wealth destruction will easily exceed wealth generation, and negative growth occurs. The goal of tax policy should be to ensure that net wealth generation is high, so we can enjoy the long term bounty of cumulative growth. This challenge is not unique to Australia – the UK has also stagnated for the better part of two decades, and for similar reasons.
This should hardly be controversial or so obscure. Taxation beyond the point of peak growth is, definitionally, picking the pocket of our grandchildren for short term benefit. Dependency is only going to continue to grow, and not just because of the National Disability Insurance Scheme. We’re aging, geriatric medicine is getting more expensive, and the economy continues to eject the highly productive sectors of manufacturing and Information/Communication Technology (ICT) in favor of axiomatically low leverage tertiary services. If government spending grows faster than the economy in general, the tax burden will continue to increase, growth will slow further, and economic crisis will result. It seems crazy to me that a well developed, productive, diverse and robust economy can be so susceptible to this sort of failure mode, but it’s happened multiple times in the last century. If we’re already taxing beyond the point of peak growth (we are) then further increases in taxation and government spending just move us further and further away from the point of maximum economic growth, digging a deeper and deeper grave for ourselves.
This graph shows the share of government spending in the Australian economy. Not all government spending is equally distortionary. It’s not particularly controversial that spending on infrastructure, defense, education, public health, and basic research is high leverage spending that pays larger dividends over time. At the same time, these sectors are beset by stagnant productivity and, not being involved in the TFP-increasing processes led by manufacturing and ICT, cannot by themselves sustain economic growth and long term prosperity.
One particularly frightening aspect of this is that inflationary spending on easily-rorted and non-cumulative social services (NDIS and similar) is the only area of the Australian economy that is actually “growing”. Call me old fashioned but GDP that counts investment in building factories is different from GDP that counts balloon sculptures at parties. Both are fun and desirable parts of a modern lifestyle, but only one of these leads to capital accumulation and can become an engine of generational wealth growth.
To make this quantitative, here’s a simple model showing that reducing net taxation by, say, 10% overall, creates the breathing room for the economy to grow, making up for short term government revenues in as little as a decade.
This graph reproduces the shape of the Laffer curve, with a peak at about 40% taxation.
This graph shows a simple Normally distributed growth curve peaking at 20% taxation. It is somewhat unrealistic as it doesn’t show negative growth for excess taxation!
This graph shows average cumulative government revenue after a number of years. The point of peak revenue moves to the left, showing that lower taxation levels are optimal for long term growth and building a foundation for a high quality of life nation.
This graph shows how the point of peak revenue shifts as a function of time horizon. As a rough rule of thumb, a 50% discounting per 25 year generation is consistent with long term economic growth and interest rates, and would suggest an optimal tax burden of around 25%, which is roughly half of Australia’s current tax rate, once you include income tax, corporate tax, GST, FBT, capital gains tax, excise and customs duties, and state taxes including stamp duty, payroll tax, and all the other times the government gets a cut.
Cutting taxes by 5%-10% sounds like a radical economic experiment, particularly for the millions (!) of Australians who depend on government pay scales for their daily bread. A sector whose low labor mobility and low economic dynamism deprives them of any chance to contribute to the private sector where, on average, their productivity will increase 6x faster.
But what’s really radical is that we have already run an unprecedented economic experiment, it has already failed quite badly, and the only solution being offered is to just do more of it.
Here are the terms:
- Drive investment away from productive enterprises by artificially restricting housing supply relative to demand. Let’s make housing so unaffordable that we end up crashing our birth rate, destroying the health of our future tax base.
- Ratchet up tax-and-spend social policies, colliding the stagnant-to-falling productivity of monopsonistic public health with the ever-rising demand of the afore-mentioned rapidly aging population.
- Increase taxes, regulations, labor law complexity, and drive the entrepreneurial class to the US (he says, writing from California).
- Identify the two key sectors whose growth is most fungible in TFP, that is, Information Communications Technology (ICT) and Manufacturing, and systematically eradicate them from the economy. Joe Walker asked Ken Henry why Australian TFP maxed out at 80% of US, and he couldn’t identify any reason other than geographic isolation from international markets, which could account for at most a 5% discrepancy. In a more recent interview, Brennan and Kaplan said something like “If you ignore ICT and manufacturing, we’re actually about as productive as anyone else.” That’s like saying that aside from cancer, trauma, infection, and aging, we expect the patient to live to the age of 10,000. Where do we expect growth to come from, if not from ICT and manufacturing, the two sectors that literally make more with less every day?
- Ignore the unearned bounty of solar PV technology for industrial uses.
We have run this experiment. It has failed. Australian growth is on the rocks. Government fiscal discipline has been all-but-abandoned. This idea that we’re enduring some hangover of high historical growth is ludicrous. If anything, past growth rate should be predictive, since the engine of technological innovation is self-improving. Growth rates have increased throughout history, from <0.1% annually during the medieval period to >2% during the industrial period to today. That’s the reason that it’s called compounding!
As if to underscore the absurdity of this basic economic fact being completely ignored (tax reform means more taxes, not less, obviously!) here is a three hour interview by Joe Walker of two leading Australian economists who both specialize in growth. What’s more remarkable is what they manage to not say, despite very probing questions and copious time.
When asked how Australia could get closer to US levels of growth, Kaplan and Brennan hemmed and hawed before saying that we could benefit from a more similar labor market, before advocating for more unemployment insurance, expansion of disability benefits, and better public health – none of which exist in any centralized sense in the US system.
What the US system does do really well is to ensure that incentives remain aligned in the economy. People who work hard increase their luck. Their income tax is relatively low, so they get to keep more of their hard-earned money, unlike in the UK where the income tax scale rises so steeply that if you don’t already have generational wealth, there’s no way to get there through honest work. The US employer doesn’t have to pay the frictional costs of the Fair Work Act 2009 so can take the risk of hiring them. Employers are free to experiment with different contract structures and as a result, work conditions in the productive, competitive, growing parts of the economy are significantly better than they are in Australia, where no such jobs even exist.
Multiple states support multiple industries, so employers and prospective employees operate in a much more robust, efficient, and dynamic labor market. Places like California are so successful at minting millionaires through employee stock options that there’s even a competitive market for risk capital funding an entire venture capital industry and thousands of experimental startups, driving a massive flywheel that generates so much wealth it spills into every corner of the globe.
Initially I thought I’d have to exclude North Korea from this global wealth spillover, but US corporate taxes fund US entitlement spending ($2.4t/year, larger than the entire Australian economy) of which over $100b/year is fraudulent, of which a significant fraction is stolen by foreign rogue state actors, of which North Korea is a global leader, likely contributing to a significant fraction of the country’s estimated $26b/year GDP. For reference, Google paid about $20b in taxes last year.
I wasn’t invited to the Australian Productivity Roundtable, but it would be unsporting of me to criticize so heavily without offering suggestions, so here goes.
What do I want?
Easier to build businesses
In the US, it’s now possible to register a Delaware C-corp (the default corporation) and set up banking, payroll, registration, and other services with a click of a few buttons. It is essential to reduce friction on the formation of companies by entrepreneurs. In Australia, forming and operating even a small business will require multiple interactions through call centers to get basic details right – a classic case of a tax-funded operation performing an unnecessary or redundant operation with low success rates, costing the citizen two or three times over.
Political representation for the interests of families and children
There’s a recognition that the Australian taxation system has contributed to bidding up the price of housing to unaffordable levels, crushing real economic growth for a generation or more. It’s really not clear how to fix this, but even making the system less unfair on the young is politically impossible if the dominant part of the electorate (home owners, landlords, and retirees) want it that way. I can’t fault them for not wanting the house they spent most of their lifetime income paying off to suddenly be worth 20% of its current value, but if we don’t make homes affordable (roughly doubling supply) then we’re all screwed long term. This outcome, of ever-increasing wealth transfers from the poor young insecure worker to the rich old secure retiree is simply a stable attractor of any democratic political system which has worked well enough to enable the survival of significant numbers of old people. At the very least, we should increase the voting power of parents in proportion to the number of children they financially support, without which children have zero representation for their interests.
Fewer, simpler taxes
Advocating for less tax doesn’t exactly take bravery, and my argument is well developed above. Income tax punishes and disincentivizes work. Payroll tax punishes hiring. Property tax punishes investment in fixed assets. Capital gains tax punishes long term investments, decreasing liquidity. Taxes on misused and underutilized capital will at least promote prosocial economic behavior, but fundamentally we must demand that the public sector find some way to increase its productivity in step with the private sector. Better competition for labor and involvement in fewer things that can be done by the private sector have worked in other countries. The other extreme is Peronism – hyperinflation and a severely bloated public sector failing to provide public services and transferring wealth from savers to politically favored constituencies.
Moderate economic inequality
Sufficiently progressive income taxation and some kind of universal basic income will delete that root of all ills, economic inequality. It will also destroy all incentives to work at all, and very quickly the productive core of the economy will grind to a halt. We’ve seen this done in the socialist and communist experiments. It did not work out, and it doesn’t take a genius to understand why. In any economic system, some people will have a natural affinity for hard work, or a natural ability at the skills that are highly prized in that decade, and they must be allowed to derive a personal benefit for that, to increase labor supply in these critical areas. Without price signals, the economy will have no way to understand how to allocate resources.
Efficient resource allocation is essential to eking out positive economic growth which, compounded over many years, raises the quality of life for everyone. How far back in the past would you be willing to go, even if you could be among the super rich? President Coolidge’s teenage son died of an infected blister in 1924, prior to antibiotics. Daniel Ludwig was the richest man in the US in 1980, but never got to use a cell phone.
It turns out that some subset of people are, for whatever reason, abnormally good at capital allocation, running businesses, or otherwise producing the positive economic outliers that drive future growth. My radical opinion is that since they are conspicuously better than the government at picking winners and generating wealth, they should be allowed to continue to do that. Note that steeply progressive income, capital gains, and wealth taxes cut against this obvious truth, and economies that pursue them quickly find out that the 500 millionaires who finally took the hint and emigrated somewhere less hostile to wealth were actually load bearing infrastructure for their entire economy. It’s not a moral black mark to be good at generating new wealth, any more than it is to be good at sport or surgery.
The optimum amount of economic inequality is not zero, just as it is not infinite. Long term, the optimum amount of wealth inequality is whatever maximizes the long term rate of economic growth. Observationally, this is a Western liberal democracy with a mixed economy, fiscal discipline, and a robust, diverse and dynamic private sector.
Reduction of dependence incentives
Societal behavior maps to incentives and vice versa. As a young man, I could never understand why the Australian hospital system was just so exquisitely underfunded that it always took four hours in Emergency to be seen by anyone. Now I’ve read a book or three about how the world actually works, I understand that any “free” public resource is typically consumed until congestion causes the marginal user to give up. The modal user of any health care system is not an otherwise healthy 20 year old with a bad cut. It’s a sick old guy convalescing in a hospital bed coping with an exacerbation of multiple interacting unfixable chronic diseases. It is literally impossible to saturate demand for free public health, and the public is understandably nervous about any officially sanctioned rationing system, including market-based price signals, so instead rationing occurs unofficially at the triage station.
More rigorously, it’s impossible to saturate demand for free anything, let alone free public health services against a backdrop of people getting exponentially sicker with age. Well and good, let’s cure aging. But at the same time, Australia has seen a massive expansion in spending under the NDIS, reaching something like 8% of total government costs, in just 10 years of operation. Who could have seen this coming, other than, you know, the NDIS architect John Walsh and the first NDIS Chair Bruce Bonyhady?
“The government should do something to help disabled people and their families.” Not controversial. “Free money for anyone with a doctor’s note.” Obviously a moral hazard. So every case needs to be adjudicated and administered, which is why 20-25% of the NDIS’s total costs are operational, which is to say for every three or four carers there is one administrator whose incentives to tell the marginal would-be that they’re not sick enough are demonstrably not working.
Caring is hard work, not particularly prestigious or well paid, and obviously vitally necessary to help our retirees maintain their dignity and independence for as long as possible. But at the same time, carers and their patients and their administrators are not inventing new technology, building factories, upgrading power plants, constructing infrastructure, or building businesses. They’re labor intensive, low leverage tertiary service jobs. They’re not driving innovation, TFP growth, or the accumulation of national wealth that will bring a better life, better options, better medicine, for future generations. A balanced and prosperous economy must be able to sustainably care for the sick, disabled, and less fortunate, but it must also care for and nourish the highly productive core economic engine that churns out enough surplus wealth to fund and to continue to fund and to continue to improve the caring aspects of our society.
16% of Australia’s six year old boys are using the NDIS, usually for cases of “mild autism”. 16%! What’s the plan when, in 14 years or so, these “disabled” kids enter the work force? Who is going to fund their entire life of dependency? Fully automated humanoid robots? From the two industries (ICT and manufacturing) that Australia, uniquely of the G20 nations, has almost no presence in?
The NDIS and other social services budgets must be capped at a level that the economy can bear, and dispensed on a needs basis. I would love a world where we can give every child and their parents UBI, but we can’t afford it yet, and we will never be able to afford it in a world where we create economic incentives for the current generation to eat the seed corn and destroy any hope of future wealth for the coming generations.
Selective immigration
I’m an immigrant. I’m sympathetic to the plight of immigrants. At times, I’ve even been pro open borders. But whatever Australia is doing right now is even less sustainable than promoting economic dependency and increasing taxation and inflation. Just as economic inequality cannot be infinite without risking social unrest, Australia’s immigration system must be oriented around ensuring the successful integration of migrants into Australia’s culture and economy.
No-one objects to migrants that work hard, pay taxes, and follow the law. But importing 500,000 people per year into a country that already has a housing crisis, is insanity. Importing 500,000 people per year, many of whom lack the education and language skills to become productive members of society, is insanity.
Cap immigration at a sustainable rate, defined by some index on housing, and thereby align incentives with housing construction. Revoke residency for the criminal or persistently economically dependent, and thereby align incentives around being a team player and building a better life. Policies that enforce pro-social behavior in migrants make it an easier sell for the next batch of migrants who would like to come to Australia.
Earlier on, I made this point earlier in a comparison on US vs Australia employment law, but the effect of making it harder to fire underperformers is not to magically make lazy or mismatched employees more productive, it’s to cut opportunities to prospective employees whose employers cannot bear the risk. Similarly, I’m in favor of the 99% of would-be migrants who will strengthen Australia’s economy and cultural fabric, which requires us to be decisive about the 1% who would ruin it for everyone else.
Again, the path forward is to align around sustainable long term growth.
Deletion of low growth ideas
At the Productivity Summit, the Australian Council of Trade Unions (ACTU) came prepared with various ideas, including:
- Reduce negative gearing (a tax benefit that allows losses on rental properties to be claimed against income tax, effectively widening the gap between mortgage and rental costs)
- Close capital gains tax breaks (which are pretty minimal in Australia anyway)
- “Mandatory enforceable agreements that would compel employers to consult with their staff” before AI is introduced to workplaces
- A levy (tax) on businesses to fund training
- A four day work week.
Imagine showing up to a Productivity Summit and suggesting that the government intervene between businesses to add friction to the adoption of AI, probably the single greatest potential growth technology ever invented.
Better industrial policy
In his interview with Joe Walker, Michael Brennan, the CEO of e61 and previously the chair of Australia’s Productivity Commission and a Deputy Secretary of the Australian Treasury, stated:
“Remember, in the year 2000 or around that time, we had a big debate in Australia about whether we wanted to be headlong in ICT—like chip manufacture—and there was a bit of a view that you had to be in that world in order to be an advanced economy. I think in retrospect, that was wrong, and Australia was right not to go down that path. But it’s really hard to know. Are there big spinoffs from being big in infrastructure like data centres? I think this is potentially an area where there’s going to be a slightly fuzzy boundary between being a developer of the technology and being an adopter of the technology.”
We’re in a bit of a pickle now. Taiwan, an island whose very sovereignty is in doubt, with a lower population and half the GDP of Australia, with no natural resources, with a legacy of destruction in WW2 and the Chinese civil war, somehow managed to secure a near monopoly on the mass manufacture of high performance microchips.
What is Australia’s excuse? Since birth, I have heard the daily mantra “It’s hard to compete with China in manufacturing.” Why? Are the Chinese blessed with better natural resources, better banking, better constitutional freedoms? No, they just work like crazy and their government realizes that national sovereignty depends on industrial power, a fact that became undeniably obvious to the Axis powers in 1942, and hasn’t changed since.
Australia invented the technology behind modern photovoltaic solar panels, and effectively paid China to take it from them.
The US and China are locked in a race to develop artificial superintelligence (ASI), an AI that is significantly better than any human at any task. The Australian government seems, at times, to understand that national sovereignty depends on having the ability to build our own submarines, guns, and drones, but seems to have blanked out entirely on the single biggest tech boost of all time – ASI. If you don’t own the weights, you don’t control the model. In a few years, Australia will have to decide whether it wants its economy to be run by a national Chinese AI or by some privately owned US AI. If the Australian government doesn’t like what Elon Musk did with Twitter/X, they’re going to really love the fact that xAI’s excellent model Grok is a leading contender to win the race to build the first superintelligence in the known universe.
Land the plane, Handmer
Tax-and-spend distorts the economy. To a limited extent, this helps promote economic growth. To a certain further extent, it buys us a society with positive attributes where the poor and sick are taken care of, at the cost of reduced growth and lower relative wealth for our grandchildren. But the distortionary effects of taxation are underestimated when we do not account for the far-from-optimal use of our finite labor supply, paid for by far-from-optimal taxes appropriated from our finite capital supply. That is, taxation hurts at least twice. It no longer surprises me that we have painted ourselves into a corner with plunging TFP growth and economic stagnation. For the sake of the long term health of our social programs and global competitiveness, we need to lift the yoke just a little and allow the economy to breathe again.