THE IMPENDING TAX CRISIS CAUSED BY AI-DRIVEN MASS UNEMPLOYMENT

There's a lot of talk about AI tech creating mass unemployment, but what does this mean for tax revenue? Is the sheer speed and scale of disruption going to cause an unprecedented crisis?

ARTIFICIAL INTELLIGENCEMACRO

Oliver Cook

4/18/20235 min read

Whichever way you look at it, the Artificial intelligence (AI) revolution is well and truly here. And, with development now going exponential, there’s an elephant in the room that no one is talking about. Tax revenue.

While there’s a growing global conversation about impending mass unemployment, few seem to look further and contemplate what that means for government revenue. Indeed, as AI technology replaces human workers across various sectors, and more high-paid jobs are lost, there will be far-reaching implications for tax revenue generation, creating a severe strain on government funding.

Too Much, Too Fast - Why the AI Revolution is Not Comparable to the Industrial Revolution

Despite high-profile warnings from the likes of Elon Musk, the inevitable truth is that as AI technology continues to evolve at a breakneck pace, automation will become more prevalent in many industries, ranging from manufacturing, logistics, and customer service, to marketing, design, and education. Jobs that were once performed by skilled humans are already being replaced by robots and AI-powered systems, resulting in displacement and unemployment for workers who are unable to compete with machines in terms of efficiency and cost-effectiveness.

While some continue to argue that automation will create new jobs, the reality is that the transition will not be smooth for many workers, especially those in middle to high-skilled jobs - i.e., the most critical taxpayers. This will result in a significant loss of tax revenue for governments. The unpalatable truth is that the AI revolution is simply happening far too quickly for our society to adapt. Illustrating this, many AI evangelists try to make analogies with historical events, like the Industrial Revolution. But, the inescapable reality is that that change happened over the course of decades, and still resulted in severe disruption and dislocation of nations and societies. And, let's not forget, during the Industrial Revolution, the population of the planet was a fraction of what it is now, and the vast majority lived in abject poverty. There was no expectation of state pensions, social services, unemployment benefits, state-provided healthcare or education, or even basic emergency services and public transport. Plus, most people didn't live much into old age! Most people were, to put it bluntly, out of sight and expendable. Such populations could be subject to far more hardships than those of today. How would Britain, for example, have fared, had the industrial revolution happened at several thousand times the speed? I would hazard a guess, ‘not very well’ - and it would have resulted in civil war, or at very least an even more stratified and divided society.

A Tax Crisis Just as Demand for Services Rockets

But, even if we ignore the individual and societal impact of mass, sudden unemployment, AI could spell disaster for tax revenue generation. By definition, high-paid jobs contribute most significantly to tax revenues through income taxes, payroll taxes, and consumption taxes. As these jobs are replaced by automation, governments will face a double whammy: not only will it lose the revenue generated by these high-paid jobs, but it will also face an increase in demand for social welfare programs, such as unemployment benefits and healthcare, to support the growing number of displaced workers. This will put a tremendous strain on government funding and will lead to a crisis in revenue generation.

Furthermore, the erosion of high-paid jobs due to AI-driven automation will have a cascading effect on the broader economy. When workers lose their jobs, they usually reduce their spending, leading to decreased consumer demand. This, in turn, will impact businesses that rely on consumer spending, leading to more business closures and reduced profits (which means reduced corporation tax revenue). As businesses suffer, they will also cut back on their workforce and reduce wages, further exacerbating the unemployment crisis and further eroding tax revenue. The cycle continues, creating a ripple effect across the economy and posing significant challenges for governments in generating tax revenue to fund essential services and programs.

Of course, elements of such cycles are seen in normal recessions and economic crises, but the difference this time is the sheer speed and scale. Just look at how far AI has come since the start of this year, and you’ll get the point. There is simply no way for businesses and educational systems, which still have to function within existing bureaucracies and comply with legacy obligations, to evolve anywhere near quickly enough. And, even with the best will in the world, there’s no way this can be addressed because every process that needs changing has to be changed via equally slow processes and systems.

The impact of AI-driven mass unemployment on tax revenue generation will likely also vary across different regions and countries. In developed countries with established social welfare programs, the cost of providing unemployment benefits, retraining programs, and other support services will be unprecedented (and that’s assuming any retraining programs can be put in place and evolve quickly enough - which seems impossible). This will lead to increased government spending and, inevitably, higher taxes, which will face resistance from taxpayers who are already struggling with a cost of living crisis. On the other hand, in developing countries with limited social safety nets, the impact of mass unemployment caused by AI can be even more severe, as governments will lack the resources to provide adequate support to displaced workers, leading to social unrest and instability.

So, at the very point when governments around the world are already in record debt, and vital services are already facing a funding crisis, tax revenue is about to crash - just when more people than ever will be needing government help. It's hard to see this as anything but a disaster.

Education Systems Facing an Impossible Task

To address the implications of AI-driven mass unemployment on tax revenue generation, governments must take proactive measures - and, they needed to take them yesterday because even today is too late. As previously mentioned, one approach is likely to be an investment in retraining and reskilling programs in an attempt to help displaced workers transition into new industries and acquire the skills needed for the jobs of the future. But, again, it is highly doubtful that any such programs will be able to identify specific targets or keep up with the pace of AI evolution. Even if they could, the massive scale of the disruption means those retrained workers will find their earnings potential decimated - which will lead to them paying less tax.

And, for anyone who thinks I’m being overly pessimistic, perhaps you should consider the education systems of the developed world (you know, the ones that are going to be tasked with ‘retraining’ the workforce). In their early teens, a student has to decide what subjects to focus on. This process of subject selection continues through college and university and takes over a decade. These subjects are taught by teachers and lecturers who themselves had to go through a decade-long process of learning and specialization. The courses they are teaching have to be proposed and researched, written, learned, and implemented - again, a multi-year process. And, let’s not forget the often multi-year process of recruiting qualified staff. The point is, this kind of thing simply can’t hope to keep pace with AI development. Not even close.

The brutal truth is that, while AI is incredible on so many levels, the world is simply not ready for it. Like it or not, our entire civilization is built on institutional foundations and infrastructure that are decades, and often centuries old. Furthermore, our biological bodies and minds are essentially the same as they were thousands of years ago. As tempting as it can be for those in the tech space to ignore the realities and invoke fanciful notions of everyone suddenly becoming hyper-productive and having more leisure time, the reality for the majority is looking distinctly dystopian rather than utopian.

Depopulation and CBDC 'conspiracy theories' seem to make more sense with each day...