Our study asked participants one simple question (followed of course by some demographic questions). We presented participants with a hypothetical scenario or vignette and asked, “Based only on the information provided, how exploitative is this scenario in your opinion?”

Now, one challenge with this approach is that many real-world examples are too entangled with existing political biases, media narratives, and emotional reactions.

If you ask whether workers in a Bangladeshi sweatshop producing trendy clothing are exploited, you will elicit predictably biased responses that align with the polarized views sketched above. Similar biases are manifest if you transparently ask about theorists’ definitions of exploitation — if you ask whether “the unequal exchange of labor is exploitation” — you will get a yes from those who accept the classic Marxist account and a no from everyone else. To eliminate various confounding factors, we constructed abstract vignettes that centered on a neutral, everyday economic activity: the production, purchase, or sale of mugs (because these are inexpensive, familiar, and morally neutral).

We then pored over existing definitions of exploitation and identified common properties that different thinkers had identified as components of exploitation. These included: inequality, whether the economic gains from the transaction were split unevenly between the parties; power, whether one party held a strong structural advantage, such as being the only available employer; basic needs, whether the terms of the transaction left the weaker party in a position where their basic human needs were unmet; disrespect, whether the dominant party displayed a disrespectful attitude toward the weaker party; and luck, whether the weaker party’s vulnerability was the result of bad brute luck, like a natural disaster. We also included injustice, a property that further distinguished power by denoting that the power advantage was not merely accidental but rather the result of a prior injustice or rights violation.

We tested all these properties in various combinations on two kinds of scenarios. Relational scenarios involved a direct transaction between two specific parties, while nonrelational scenarios described broader patterns of how wealth and resources are distributed across society.

Finally, we developed two baseline cases that excluded any potentially exploitative factors: one relational and one nonrelational. In total, this gave us thirty-one clear vignettes to test. This modular design allowed us to isolate the exact impact, or “treatment effect,” of each distinct property on participants’ evaluations.

Who were our participants? They included 551 professional academic philosophers recruited from top-rated global institutions and 1,546 members of the general public, split pretty much evenly between participants based in the United States (777) and the United Kingdom (769). The participants read through 10 vignettes, one at a time. They were first shown the two baseline vignettes and then eight more from the pool of 31.

Here’s an example of a relational vignette that includes the properties of power, inequality, and injustice:

B wants mugs and A is the only vendor selling mugs because other sellers have been unjustly excluded from the market. A sells mugs to B and the agreed price benefits A more than B.

For each vignette, participants answered the same question: “Based only on the information provided, how exploitative is this scenario in your opinion?” They recorded their judgments using an active slider scale ranging from 0 (“not at all”) to 50 (“moderately”) up to 100 (“maximally”).

One possible concern with this approach is that a score of 35 might mean different things to different participants. We address this concern by using “individual fixed effects” in our baseline empirical model, which allows us to control for constant but unobserved influences on individual evaluations, including differences in “exploitation anchors.”

In practical terms, we estimate how much the presence of power, for example, changes the average reported score for each individual, and then calculate the average of these differences across all individuals. Importantly, the results are much the same when these fixed effects are excluded or observable characteristics are included instead.

So, what did we find? The results were rather surprising. It turns out that experts and members of the public order the properties associated with exploitation in nearly the same way. Neither group thought being relational/nonrelational or the presence of bad luck made much of a difference to exploitation. The scores for both were close to zero. Here’s a table showing the order and scores for the properties that made more of a difference for both participant pools, all of which are highly statistically significant.

As you can see, the two participant groups only differ in the relative importance they attach to inequality and power, with experts rating power over inequality and the public ranking them the other way around. And now to the result teased in the introduction: when we divided the two groups down by political orientation, the ordering of the properties was the same. Liberal and less liberal academics ranked the properties in the same order shown in the table. The same is true for more and less liberal members of the public. Further, we asked philosophers whether they identified as libertarians, communitarians, egalitarians, capitalists, or socialists. Again, the ordering was the same. At least on this issue, libertarians and capitalists agree with socialists and egalitarians: injustice, inequality, and power are the most significant drivers of exploitation.

Now, it is important not to overstate the similarities between these groups. Participants’ sociopolitical orientation significantly impacted how much exploitation they saw. For example, libertarians’ scores were lower across the board. Compared to egalitarians they saw less exploitation in vignettes containing inequality (-10) and power (-15). Nevertheless, the fact that groups associated with the Left and Right gave the same orderings for the properties mean that both agree about what exploitation is. The lower scores from groups on the Right mean they are less likely to say that any particular scenario is exploitative.

There’s one more result that’s worth emphasizing. While power alone, and inequality alone, move the needle for participants, the presence of both properties increases scores more than the sum of their individual effect.

For example, for academics, vignettes containing only power raised scores by +14 and those with only inequality by +8, but when both appeared together scores increased by +34, which is 12 points greater than the sum of both alone. This “interaction premium” suggests that exploitation is more than the sum of its parts. This may be the study’s most important finding because it shows that the core of exploitation arises from a combination of factors. What triggers the strongest moral condemnation is the leverage of unequal power to engineer unequal outcomes.

In other words, exploitation is recognized as effective bullying — the conversion of a structural power advantage into an unfair share of the economic pie. This result overturns two popular approaches to exploitation in the academic literature: purely distributive approaches, which take exploitation to merely be the maldistribution of gains, and purely power-based approaches, which argue that exploitation arises solely from the dominating use of power.

First, as we underscored in the introduction, our results reveal that exploitation is not an arbitrary or highly partisan invention. It is a stable, deeply embedded, and shared social concept. Across different classes, political outlooks, and educational backgrounds, people use a shared moral framework to evaluate exploitation: we look at the historical context, evaluate the balance of power, and measure the equity of the outcome.

While agreement between the Left and Right on exploitation is surprising — especially given the contrasting views of Nozick and Marx presented above — perhaps it should not shock us altogether. After all, some of the populist right’s core anxieties arise from concerns about the cost of living, precarious employment, and the dominance of “big” industries (such as tech, pharma, and finance). And these overlap with perennial critiques from the political left. At their root, both the Left and Right are responding to the same structural reality: the deteriorating material conditions and economic vulnerability of large segments of the population.

Second, the contemporary Left and Right — or at least their populist iterations — share another feature: a concern with political correctness, albeit in different forms. This concern is perhaps more charitably interpreted as a concern about esteem — or respect.

While respect is an important moral value, an overemphasis on interpersonal attitudes and manners can shift the conversation away from hard structural metrics like wages and workplace power dynamics. To align with the current grievances, corporations pour resources into branding and public relations, changing their message to suit different markets. These attempts to humanize their operations by emphasizing themes of respect, inclusivity, and empathy in their corporate culture move the ethical conversation away from the material and power dynamics that really drive exploitation.

Our study reveals a sharp divergence between experts and the public on the value of respect, shedding light on a feature of modern alienation. The disrespect variable had a negligible impact on academics’ scores (+2) — a polite boss, after all, extracts value just as effectively as a rude one. But for the general public, disrespect carried real weight (+12).

This gap helps explain a crucial feature of our current political climate. Public discourse has focused on the increasing feelings of “alienation” and “loss of voice” of large segments of the population over the past few decades. One interpretation of our results is that this increased emphasis across the political spectrum means that for many working people, interpersonal disrespect is not just a breach of etiquette; it is the immediate, experiential manifestation of their structural vulnerability. It is how asymmetric power is lived and felt — whether on the shop floor, in the gig economy, or at the rental office.

The agreement between the Left and Right — and indeed across many other demographic dimensions — that exploitation is the weaponization of unjust power to secure unequal outcomes suggests a path forward for the Left’s political agenda. It is clear that the public is fed up with the ideological defenses of low wages and bad options for work, housing, and health that appeal to the language of consumer choice and flexibility.

If we want an economy free from exploitation that speaks to the material concerns that motivate populist politics on the Left and Right alike, then we must focus on policies that deliberately dismantle power asymmetries and protect vulnerable actors from being forced into desperate terms. That means increasing the power of vulnerable workers, which will require sectoral bargaining mandates to revitalize organized labor, increased minimum wages, and universal public housing and health care. And decreasing the power of the capitalist class, which will require wealth taxes and increased antitrust enforcement and regulation. When power is balanced, the capacity for one side to extract highly unequal, exploitative gains is dramatically curbed.

Let us be clear. The message here is not that those on the Left should moderate their policies in the name of bipartisan harmony. The view of exploitation that emerges from our study is not that the truth lies somewhere between Marx and Nozick. Rather, the headline is that those on the Left and Right agree on a view of exploitation that is located much closer to Marx’s view, one that emphasizes the use of power accumulated through injustice to extract unequal gains.

The interpretation of these results that we want to stress is that they represent an opportunity for the Left. They provide further evidence that the motivations of many populist views across the political spectrum are in fact compatible with historically leftist positions on power and equality.

While contemporary headlines remain dominated by the culture wars, they frequently mask these deep-seated economic anxieties. The corporate preoccupation with etiquette and notions of respect that are not linked to power operates as a tempting distraction — one to which both the Right and Left are susceptible — from workers’ own anti-capitalist intuitions. Our study shows that a structural understanding of exploitation is not a niche theoretical framework but a commonsense shared reality. Rather than playing defense on the terms set by corporate public relations, the Left has a distinct opportunity to reanchor the political conversation around shared material interests and policies that dismantle the very power asymmetries that the public already condemns.