AI and automation are reviving old economic structures ruled by inequality. Household servants – maids, couriers, pet carers and food delivery workers – are being reborn behind the convenient guise of the gig economy. Astrid Krenz and Holger Strulik write that this is not a cultural phenomenon, but a predictable outcome of structural economic forces such as automation, inequality and shifts in high earners’ time allocation decisions.
From factory robots to generative AI, the accelerating pace of automation is reshaping the future of work. The developments are changing the labour market in ways that affect all of us. If you are thinking about the future for your children, what kind of education or career path would you recommend?
The answer is far from obvious.
Much attention has focused on how new technologies threaten traditional jobs. However, less has been said about the broader economic structures they are reawakening. Rising inequality in the age of automation appears to be reviving a feature of the past many thought long gone: the servant economy.
In a world where algorithms write emails and robots build cars, household service jobs—maids, couriers, pet carers, and food delivery workers—are growing rapidly. These roles, once on the decline, now form a central part of what is often referred to as the gig economy. But beneath this surface-level convenience lies a deeper story about how technological change and inequality are transforming labour markets in more traditional ways than we might expect.
Outsourcing domestic work
In our recent paper, we develop a macroeconomic model to explore how households divide domestic tasks between self-performance and outsourcing, and how this allocation responds to automation both at home and at work. By calibrating the model to US data, we trace how technological shifts have influenced household time use and the employment structure of domestic service from 1900 to 2020. Through a series of counterfactual simulations, we explore the roles of automation, the emergence of new domestic tasks, and the gig economy in shaping these trends.
Our study reveals a striking historical symmetry. In the early 20th century, domestic service was widespread: in 1900, nearly 12 per cent of US households employed servants. But this began to change as appliances such as washing machines, refrigerators and vacuum cleaners revolutionised domestic life.
Alongside rising access to education and declining income inequality, the share of households employing servants plummeted to just 3.5 per cent by 1950. The servant economy, it seemed, was a relic of the past.
Firm automation
But from the 1970s onwards, a reversal began. The trend towards outsourcing domestic tasks was initiated and propelled by skill-biased technological progress at the workplace. Innovations in firm automation, rather than in households, increased productivity and earnings for high-skilled workers but not for those with lower skill levels, implying a rising skill premium.
As inequality rose, so did the opportunity cost of time for high-income individuals, prompting them to outsource an increasing share of domestic tasks to lower-skilled workers, “transforming wealth into happiness”. The saved time from household chores can be spent on wage work, leisure or new time-intensive tasks of home production such as caring for the elderly and the children, and their entry into prestigious colleges.
We are witnessing a return to the servant economy in a 21st-century version— one shaped not by domestic necessity, but by the time constraints and wealth of a growing professional class.
Our research shows that this return of the servant economy is not a cultural coincidence. It’s a predictable outcome of structural economic forces: automation, inequality and shifting time allocation decisions by high earners.
Efficient outsourcing
To test our model empirically, we analysed panel data from US metropolitan areas between 2005 and 2020. We found that an increase of 1,000 industrial robots leads to a 5.1 per cent rise in inequality, measured by the 90–50 percentile ratio of wages. In turn, higher inequality fuels demand for domestic services. A 10 per cent increase in inequality is associated with a 2.9 per cent rise in employment of maids, 8.3 per cent for couriers, and around five per cent for animal caretakers and for an aggregate of services jobs. Wages in these roles also increase in response to rising inequality.
We also captured the dynamics of the gig economy. Platforms that match service providers to households have significantly lowered transaction and coordination costs, enhancing the efficiency of outsourced domestic work. This technological improvement boosts demand for such services and interestingly, it also helps moderate the growth of inequality by raising wages for these workers.
From a policy perspective, our findings reinforce the enduring value of education. Consistent with the canonical model of the skill premium, we find that a greater share of college-educated workers reduces inequality.
Shaped by the future
What is striking is that in many ways, we are witnessing a return to the servant economy in a 21st-century version— one shaped not by domestic necessity, but by the time constraints and wealth of a growing professional class. The tools have changed, but the underlying dynamics echo the past.
Understanding this shift is critical for business leaders and policymakers alike. As automation and AI continue to evolve, their impact on labour markets will go well beyond which jobs disappear. They will determine how economic roles are redistributed, how inequality evolves, and how time itself is valued.
Sign up for our weekly newsletter here.
- This blog post is based on Automation and the fall and rise of the servant economy.
- The post represents the views of its author(s), not the position of LSE Business Review or the London School of Economics and Political Science.
- Featured image: Kitchen Scene, by Peter Wtewael (1620s), courtesy of the Metropolitan Museum of Art (New York), Open Access (Public Domain)
- When you leave a comment, you’re agreeing to our Comment Policy.
This is overstated. The gig economy in the US is no more than 1 to 2 per cent of the workforce according to the US BLS. It is not clear how fast it is growing, but it is not at the expense of “traditional jobs” or greater wage inequality because it is mostly replacing one form of self-employment with another according to the OECD. Overall, there has been no significant shift in either wage inequality or employment structure in most OECD economies over the period covered by the study, including the UK. We are not heading to the return to a “servant economy”. It is questionable that courier deliveries or pet walkers are part of a “servant economy” in the same way as, say, maids or other domestic servants. We are seeing growth in services linked to household consumption, but that has been going on for many decades and there is nothing very new about it. I fear that marginal changes are being given a catchy but inaccurate label to give them an importance they do not deserve.