When algorithms displace workers, who supports them, and how do governments pay for it? Consider Amazon. The company now operates hundreds of thousands of robots in its warehouses, including new AI-enabled systems capable of sorting, lifting and moving goods with minimal human input. The company linked the reduction of 14,000 roles in October last year to AI-driven innovation. If AI shifts value from labour to capital, governments will have to rethink both how they tax the economy and how they help workers adjust. 

Across OECD countries, about a third of the cost of employing a worker goes to taxes and social-security contributions, generating close to half of all tax revenues. Technology, by contrast, is rarely taxed. It is often subsidised through accelerated depreciation, investment allowances and other incentives designed to encourage firms to adopt new tools and remain competitive.

This asymmetry reflects a longstanding policy logic dating back at least to the 1980s: countries want firms to invest in productivity-enhancing technologies. Businesses that fail to adopt new technologies tend to lose market share over time. Hence, proposals such as a “robot tax” or taxes on computing power have struggled to gain political support. Governments fear that taxing technology directly would slow innovation and weaken competitiveness.

For this reason, governments should prepare now for an AI-driven shift away from labour income. Two priorities stand out: broaden the tax base beyond labour and expand the institutions that help workers adjust to technological change. 

Changing times

Over the past half century, technological investment has gradually reduced labour’s share of income (Karabarbounis & Neiman, 2014). In most advanced economies, the decline has been modest, and the adjustment manageable. Productivity gains expanded overall output, so that even as labour’s share fell, the total wage bill continued to grow.

Artificial intelligence could change this. Unlike earlier general-purpose technologies such as computers or the internet, AI is diffusing more rapidly and can perform a wide range of cognitive tasks that were previously beyond the reach of machines. When embedded in robotics, it can also expand the automation of manual work.

AI does not need to eliminate most jobs to create a policy crisis. It is enough if it generates significant mismatches in the labour market. These mismatches may emerge along dimensions such as skills, seniority, geography and time. The jobs AI creates or complements may require very different skills and levels of experience from those it replaces. They may also be concentrated in different regions. And new jobs may emerge more slowly than old ones disappear. Even if the most disruptive scenarios never fully materialise, the possibility of such mismatches should prompt governments to prepare for a larger and faster adjustment.

How should tax policy respond?

Expanding fiscal capacity does not mean taxing everything other than labour. New revenue sources should ideally be broad-based, minimise distortions to innovation and investment, and remain simple to administer.

A common instinct is to tax capital more heavily as AI shifts income away from labour. There is a strong case for doing so, particularly when profits reflect economic rents rather than productive investment. Much of the AI revolution relies on public inputs, from publicly funded research and telecommunications infrastructure to vast amounts of user data. Allowing these gains to accrue entirely to private shareholders risks what economist Mariana Mazzucato has described as the socialisation of risk and the privatisation of rewards.

In practice, however, taxing capital is difficult. Governments struggle to distinguish normal profits from excess profits, and rising capital mobility has fuelled tax competition. Across Europe, the average corporate tax rate fell from around 50% in the mid-1980s to just over 20% today. At the same time, many of the most profitable digital firms operate globally while generating revenue from users in many different countries.

One response has been the introduction of digital services taxes (DSTs), which allow governments to tax revenues generated by digital platforms from local users. Several European countries have already adopted such taxes, targeting advertising and online marketplace services. Although technically levied on revenue, DSTs can reduce returns to shareholders when firms cannot fully pass the tax on to consumers, effectively acting as a proxy for taxing digital rents.

But capital taxation alone is unlikely to provide a stable fiscal foundation. As labour’s share declines, governments may also need to rely more on consumption taxes. Value-added tax is particularly attractive because it does not discourage investment. When paired with redistributive transfers, as in New Zealand’s Goods and Services Tax system, it can also be made progressive, meaning that lower-income households receive more in transfers than they pay through the tax.

The challenge, then, is not simply to tax capital more, but to rebalance tax systems that currently rely too heavily on labour income. A mix of rent taxation and broader consumption taxes may provide the most robust fiscal base in an AI-driven economy. Fiscal capacity, however, is only the beginning. The real challenge is how governments use these resources to support workers through technological disruption.

Adjustment architecture 

The automation wave of the past four decades left many regions in Europe and North America facing economic insecurity, declining status and political polarisation. Yet countries responded very differently. Evidence suggests that those with stronger redistribution and more active labour-market policies weathered the shock far better. Inequality, health outcomes and political polarisation deteriorated far more in the United States and the United Kingdom than in much of continental Europe and Scandinavia.

Among the countries that adjusted more successfully, different institutional models exist. In Germany, adjustment largely happens within firms through short-time work schemes, wage moderation and coordinated retraining. Denmark’s “flexicurity” model instead facilitates adjustment between firms, combining flexible hiring and firing with generous unemployment benefits and strong retraining programmes.

Despite these differences, these systems share one feature: a strong worker voice in labour-market institutions. In Germany, this takes the form of co-determination, works councils and sectoral bargaining. In Denmark, union coverage remains high and wage bargaining is coordinated. By contrast, the United States and the United Kingdom largely treat job displacement as cyclical unemployment rather than a structural transition requiring active adjustment.

Even countries with well-developed adjustment institutions will need to prepare for faster disruption. One promising but still underdeveloped tool is the individual learning account: a dedicated training fund that workers accumulate over time and can draw on to finance retraining and income support during periods of study. France’s Compte personnel de formation and Austria’s paid training leave (Weiterbildungszeit) offer examples of how such systems might work. 

Another concern is the vulnerability of entry-level cognitive jobs to AI. Even well-equipped countries may therefore need to expand programmes that help young workers establish early connections with employers, including internships and apprenticeship-style training programmes.

Countries with weaker adjustment institutions may need a different approach if they cannot build such systems quickly. An idea discussed in Korinek and Lockwood (2026) and in line with Mazzucato’s socialisation of rewards is to give citizens a direct stake in national wealth through a universal basic capital scheme administered through a sovereign wealth fund. Many such funds rely on capturing specific economic rents, for example from natural resources or, potentially, from AI. But a similar institution could also operate as a public saving mechanism, gradually converting part of general tax revenue into collectively owned capital. Over time, such funds could distribute dividends while steadily building a shared national asset base.

The profound transformation of our economies by artificial intelligence is already underway. It could bring dramatic gains in prosperity, but also rising inequality and social and political upheaval. The institutions that tax economic activity and help workers adjust were designed for a world in which labour played a larger role than capital. Updating them will not slow technological change, but it will help determine whether the benefits of AI are broadly shared or narrowly concentrated. The cost of preparing may be significant, but the cost of failing to prepare could prove far greater.

References

Boháček, R., Bueren, J., Crespo, L., Mira, P., & Pijoan‐Mas, J. (2021). Inequality in life expectancies across Europe and the US. Health Economics, 30(8), 1871–1885.

Korinek, A., & Lockwood, L. M. (2026). Public finance in the age of AI: A primer (NBER Working Paper No. 34873). National Bureau of Economic Research. https://doi.org/10.3386/w34873 

Brookings Institution. (n.d.). Does automation increase support for the far right? https://www.brookings.edu/articles/does-automation-increase-support-for-the-far-right/

Karabarbounis, L., & Neiman, B. (2014). The global decline of the labor share. Quarterly Journal of Economics, 129(1), 61–103.

Mazzucato, M. (2013). The entrepreneurial state: Debunking public vs. private sector myths. Anthem Press. 

OECD. (n.d.-a). Individual learning schemes. https://www.oecd.org/en/publications/individual-learning-schemes_203b21a8-en.html

OECD. (n.d.-b). Tax wedge (indicator). https://www.oecd.org/en/data/indicators/tax-wedge.html

OECD. (2025). Revenue statistics 2025. OECD Publishing. https://doi.org/10.1787/3a264267-en  

Parolin, Z. (2021). Automation and occupational wage trends: What role for unions and collective bargaining? OECD Social, Employment and Migration Working Papers, No. 256. OECD Publishing. https://doi.org/10.1787/596b32ce-en 

We are living through one of the most disruptive technological transformations in history. AI, robots, and their interaction are reshaping workplaces at a pace unmatched by previous technological disruptions. Bold predictions about how AI may reshape work are likely not exaggerated. Yet, we don’t hear much from the policy front. As an economist with an avid interest in public policy, I have struggled to understand the role of policy in steering the relationship between technology and labor. Should policy intervene at all, and if it does, at what level should this be? Should it create incentives that direct firms’ innovation away from labor substitution and towards labor augmenting technologies? Should it reimagine education such that the investments in human capital can compete with the investments firms make in technology? Or should we tax technology the way we tax labor?

Economics teaches that if a technology can perform a job more cost-effectively than a human, firms will eventually adopt it. Firms have little choice: if they do not adopt, but others do, the cost of their goods or services will not fall relative to that of adopters and they will see themselves being outcompeted. In the long run, only the adopters survive. 

History offers plenty of evidence. Agriculture used to employ up to 80 percent of total labor in many countries at the turn of the 19th century. Mechanization and logistics brought this down to 3 percent or less across the Global North today. Manufacturing used to peak at above 30 percent of total employment; now it is down to 15 percent in some economies as a result of mechanization and robotization. Clerical work peaked at 20 percent just to see itself reduced by half of that by computerization.

These dynamics play out globally as well - countries that do not adopt lose out to adopters and risk the loss in global competitiveness. This is why governments typically choose to subsidize, rather than tax, technological innovation and adoption through schemes like accelerated depreciation or investment allowances. Faster adoption means greater competitiveness, nationally and globally. There are more perks to adoption - technology increases the wages of workers whose work it complements. And then there is the downside - adopters hire less of the workers whose labor technology substitutes. 

If we cannot tax technology, what else can policy do? Historically, technology has simultaneously destroyed and created jobs. The policy challenge therefore hasn’t been the one of solving mass unemployment, but one of relocating workers from obsolete to new jobs. Mobility subsidies can help workers move from regions in decline to flourishing regions, reducing geographic mismatches in job supply. Systems of education offer opportunities for training and retraining, reducing skill mismatch. 

Can the same policy playbook still work today? Today’s technological transformation is different. First, the pace of change is greater - compared to other general purpose technologies like electricity, computers or the internet, AI is diffusing faster. Second, the technology is more powerful - the breadth of human tasks that AI, robots and their combination can do is greater than that of past technologies. It can perform more complex cognitive tasks than any past technology. Third, the jobs it creates may be distant from the jobs it destroys, temporally and skills-wise. If the pace of change is fast enough, one could imagine a world in which many jobs disappear before new ones emerge. The currently emerging jobs — AI prompt engineer, data engineer, ethicist — may not be accessible to everyone.

If we cannot tax technology, can we tax labor less? The tax wedge, that is the share of tax in total labor cost, neared 35 percent on average across OECD countries in 2024. This puts a great deal of incentives for firms to invest in technology rather than labor when labor-saving technology is available. Some of this, I would argue, could be put in service of workers. One direction for policy would be to direct some of this burden on labor towards financing individual training accounts - dedicated training funds that individuals accumulate over time and can use to pay for re-training and living expenses during periods of training. In France, for instance, a payroll-based training contribution by employers funds the French Individual Learning Accounts (Compte personnel de formation). In Austria, some of the tax on labor is redistributed back into adult education through Austria’s paid education leave (Bildungskarenz), where employees can take up to 12 months of paid leave to spend on training. Such schemes can help bridge the skill gap that technology causes.

How about impacting the direction of technological innovation through policy? Technology substitutes some job tasks, and augments others. Each of these increases productivity - the first only by reducing the need for labor, the second by making labor more productive. So why don’t we subsidize the innovation and adoption of labor augmenting technology? It’s worthwhile to spend some time thinking about this idea even if the devil might be in the details: augmentation may be difficult to measure and firms may find it all too easy to label tech as labor-augmenting. 

But what if the transformation is disruptive enough to create a temporal and skill gap that individuals find difficult to bridge? This is a scenario that policy rarely tackles, but the time might be ripe to consider the options. Imagine a world where, at least temporarily, most work in an economy is done by technology, and the work that is done by humans requires extreme levels of skill. The current fiscal systems that rely heavily on labor taxation will collapse, and there will be little to redistribute in order to support the many jobless and ‘unskilled’ people. To maintain a fiscal system that can keep investing in humans, the source of fiscal income will need to change from labor to capital. 

One idea is to take a public stake in AI. Building AI requires public infrastructure and vast amounts of citizens’ data, and governments could charge data use fees, for instance. AI is expected to generate rents - large profits due to network effects, scale, and control of data. This is similar to oil production in a sense, and the way governments take a public stake in AI rents and productivity can be modeled after the idea of sovereign wealth funds in oil-rich economies, e.g., Norway’s Government Pension Fund Global. Different from traditional  sovereign wealth funds, the fund could finance socially valuable work that the market under-values - childcare, elderly care, community engagement, and environmental protection. In this way, policy does not only help protect livelihoods, but also helps maintain the critical value that work provides.

Creative destruction has always been a feature of technological progress. This time, the transformation is faster and more sweeping than ever before. We see the Red Queen effect unfolding before our eyes: to keep pace with technology, humans must run faster than ever. Our most powerful lever is public policy and we should not shy away from using it.