The word “inequality” brings to mind all the trappings of modern-day wealth and luxury: mansions, yachts, private jets and secluded islands. But the roots of that inequality have a distinctly different character, according to a new study.

Bovines — oxen, in particular — were the chief drivers of early inequality, according to research published in the journal Antiquity, which examined the archaeological records left by 150 ancient societies on both sides of the Atlantic. The findings rely, in part, on new methods to measure wealth inequality across societies at different points in their development.

The traditional narrative of early inequality holds that the development of agriculture, which allowed individuals to store large stocks of surplus grains and other goods, was the prime mover of the near-universal human divide between the haves and have-nots. But the research expands on and confirms an earlier, smaller study’s finding that inequality didn’t really begin taking hold until thousands of years after the development of agriculture. That ignition point happened right as the domestication of oxen was becoming widespread, around 4,000 B.C.

Oxen, in this new telling, revolutionized farming by allowing people to work much larger plots of land. Before, farmers had to till soil by hand using hoes or other simple tools. But an ox, harnessed to a plow, could do the same amount of work in a fraction of the time.

Study author Amy Bogaard of Oxford University estimates that before the introduction of the ox-driven plow, the typical antiquity-era family might have been able to manage a small, “garden-style” farm of about 1 hectare, or 2.5 acres. But add oxen to handle the heavy work of turning over the soil for planting, a single family’s output “would be multiplied by a factor of 2, 5 or even 10-plus,” depending on things like the condition of the land and of the animals.

Oxen were, in other words, one of the first forms of capital — an asset that could generate economic value for its owners. The study’s authors liken them to the factory robots used today: “a form of labor-saving technology that led to a decoupling of wealth from labor — a decoupling fundamental to modern wealth inequality.”

The more oxen you had, the more land you could farm — and the more land you farmed, the more oxen you could afford, in the Neolithic version of modern-day capitalism’s positive feedback loop.

One of the smoking-gun pieces of evidence in favor of the oxen hypothesis is the difference in inequality among ancient societies in Europe and Eurasia, where oxen were widespread, and those across the Atlantic in the Americas, where the animals weren’t introduced until the time of Christopher Columbus. Those pre-Columbian societies had no equivalent beasts of burden capable of handling heavy agricultural work, and the new study found inequality in those societies was typically lower than in the Old World.

To make such comparisons possible, the authors analyzed four types of household wealth that are visible in the archaeological record: land, household storage space, household living space and goods buried in the graves of the deceased. This record is by necessity broken and incomplete: Certain goods decayed over time, were destroyed or were stolen. Certain individuals, such as wealthy tribal leaders or priests, were more likely to leave an archaeological footprint than a common laborer who dies penniless.

Much of the focus of the research was in correcting, to the best extent possible, the lopsided biases inherent in the record. That was done, in part, by examining relationships present in some of the most complete sites with the most comprehensive records of habitation, as well as in some early modern sites, like 15th-century Florence and parts of 17th-century Germany.

The end result isn’t perfect — blank spots in the historical record can never be truly filled in — but the authors write that their estimates comport with earlier estimates, using a different methodology, published in 2017 in the journal Nature.

“If there are opportunities to monopolize land or other key assets in a production system, people will,” Bogaard said in a statement. “And if there aren’t institutional or other redistributive mechanisms, inequality is always where we’re going to end up.”