The Joneses Visit an Economics Lab

Why do some people spend money on luxury goods simply to signal status, while others seem to spend merely to avoid feeling inferior to their neighbors? This question sits at the heart of behavioral economics. It explores how social comparison shapes our consumption. Researchers have long debated whether we are driven to climb the social ladder or to avoid falling behind it.

A new study from the University of Essex proposes that neither impulse acts in isolation. Instead, the researchers suggest that spending is governed by a specific psychological architecture. We experience a sharper emotional sting when we fall behind our peers (envy) than we do joy when we surpass them (pride). By testing this in a controlled laboratory environment, the authors find that this "loss-averse" social comparison is the most accurate predictor of behavior.

The failure of simple social models

For decades, economists have relied on models to explain "conspicuous consumption"—spending on goods that serve primarily to signal status. Some theories suggest a "Keeping up with the Joneses" effect. In this model, individuals aim to match the average consumption of their peer group. Others propose "rank-based" models. In these, the goal is to secure a specific position in the social hierarchy, much like a player in a tournament aiming for a top-three finish.

However, these models often struggle to account for the nuances of human emotion. A simple average-based model fails to explain why people react differently to a peer who is slightly ahead versus one who is significantly ahead. Similarly, rank-based models predict that people will spend the absolute minimum required to hit a target rank. Yet the authors find that real-world spending is rarely that surgically precise. Previous attempts to bridge this gap lacked a unified framework to distinguish between these competing psychological drivers.

Modeling envy and pride through Prospect Theory

To resolve this, the authors developed a common mathematical framework to compare seven different models of social preference. Their core innovation involves adapting "Prospect Theory"—a framework originally designed to explain how people make decisions under risk—to social comparison. In Prospect Theory, individuals do not view gains and losses symmetrically. Losing \$100 feels much worse than winning \$100 feels good. This is known as loss aversion.

The researchers applied this logic to social standing. They proposed the Prospect Theory - Peerwise Envy-Pride (PT-PEP) model. This model treats falling behind a peer as a "loss" and surpassing a peer as a "gain." As shown in, the model utilizes an S-shaped value function.

This function features a "kink" at the point of zero difference. This kink represents the asymmetric weight humans place on envy versus pride.

The mechanism works through pairwise comparisons. Rather than looking at a single group average, the model calculates utility by comparing an individual's consumption to every single peer in their group. Crucially, the model incorporates "diminishing sensitivity." This means the marginal emotional impact of a peer's spending decreases as the gap between them grows wider. This prevents the model from predicting the extreme, unrealistic spending spikes seen in simpler, linear models.

Evidence from the decal experiment

To test these theories, the authors conducted a laboratory experiment involving 224 participants. Players earned tokens through a repetitive manual task, similar to a "slider" interface .

They then chose how to spend those tokens on digital "decals"—purely decorative items that vanished after each round. The experiment varied visibility. In the "Jones" treatment, peers were anonymous. In the "Veblen" treatment, players could see exactly who was spending what.

The authors report several striking findings. First, conspicuous consumption (CC) responded almost one-for-one to changes in the average spending of peers. Second, visibility mattered immensely. CC was higher and more persistent in the Veblen treatment, where social signaling was possible. Most importantly, when the authors allowed players to act as "last movers"—giving them a chance to revise their spending after seeing what others did—per capita CC jumped by 2.04 tokens .

When analyzing the data to find the best model fit, the authors found that the PT-PEP model outperformed all others. Specifically, the paper finds that CC decreases as the "dispersion" (the spread or inequality) of peer spending increases. While a simple average model would predict that a more diverse range of peer spending wouldn't change a target, the PT-PEP model correctly predicts that as peers' spending becomes more scattered, the urge to compete actually diminishes.

Limitations of the digital playground

While the results are compelling, the study has notable boundaries. The researchers acknowledge that "decals" are a relatively bland form of conspicuous consumption. Unlike a luxury car or a designer handbag, a digital sticker lacks complex cultural baggage. This might lead to a more sterilized version of social competition than what occurs in actual markets.

Furthermore, the "Last Mover" effect requires cautious interpretation. The authors note that it is unclear if this jump is a purely strategic adjustment to peer behavior. It could also be a "restart effect," where the new phase of the experiment simply re-energizes participant engagement. Finally, because the income was earned through a simple mechanical task, the experiment may not fully capture how people signal complex traits like intelligence or professional success.

The verdict: A loss-averse social engine

The PT-PEP model provides a robust, mathematically grounded framework. It successfully navigates the messy reality of human emotion. The study demonstrates that social comparison is fundamentally asymmetric. It is also sensitive to the distribution of the group. The authors have provided a much more sophisticated toolkit than the blunt instruments of "averages" or "ranks."

For researchers looking to simulate social dynamics, the takeaway is clear. Stop treating social status as a simple linear scale. To accurately predict how people will react to rising inequality or shifting social norms, one must account for the fact that the sting of being left behind is far more powerful than the satisfaction of being ahead.

Figures from the paper