2026 · American Economic Review
A Model of the Attention Economy: Data, Platforms, and Discovery
Verdict
UNSCIENTIFIC
Neither the paradigm nor the study inside it passes. The verdict is unscientific.
1·Study methodology · inside the paradigm
UNSCIENTIFIC
Does not follow the scientific method within Neoclassical Economics.
2·Paradigm · Neoclassical Economics
UNSCIENTIFIC
Fails the tests. The worse layer decides.
How the verdict is decided
The verdict grades both layers: the paradigm this study assumes, and how the study was carried out inside it. Both must be scientific for the whole thing to be scientific. A clean method inside an unscientific paradigm is unscientific. An unscientific method inside a better paradigm is unscientific. A split on either layer is mixed. The worse layer decides.
1·Study methodology
Did this study test its claim with methods that are independent, falsifiable, and non-circular?
UNSCIENTIFIC
This study claims that free platforms and targeted ads can be modeled so that data policies reverse effects over time, platform quality is too high or too low, and ad revenue shrinks relative to GDP. All three tests fail.
Independently Verifiable
Fail
No one recorded anything: the awareness, attention and preferences the equations run on were typed in by the author, not observed.
Falsifiable
Fail
Effects that can reverse, and quality that can be too high or too low, leave no outcome that would count against the model.
Non-Circular
Fail
Limited awareness and ad-driven discovery are assumed at the start, then reported at the end as what the model found.
Why
The author wrote down a set of assumptions: consumers only know about some firms, platforms pay to build free services, and platforms sell ads aimed at what they know about each user. He worked out what those assumptions produce over time. The outputs are statements like 'effects can reverse' and 'quality may be too high or too low.' No people, platforms or ad budgets were measured. The one real-world pattern mentioned, ad revenue falling against GDP, is something the model is built to be able to reproduce. Whatever happens in the world, some setting of the assumptions can match it, so the model was never put at risk.
2·Paradigm · Neoclassical Economics
Does the framework this study assumes pass the three tests?
UNSCIENTIFIC
Independently Verifiable
Fail
Preferences, awareness and welfare are never observed. Only purchases and ad dollars are visible, and the framework reads preferences into them.
Falsifiable
Fail
A market that misbehaves gets called a friction, an externality or an inefficiency, so no market outcome counts as a disproof of the framework.
Non-Circular
Fail
Preferences are inferred from what people chose, then used to explain why they chose it.
Why
Neoclassical Economics explains behavior by preferences and optimizing choices. Nobody can watch a preference. What anyone can see is people buying things, platforms charging for ads, and money changing hands. The framework infers the preferences from those choices and then uses them to explain the same choices. When markets do not behave as predicted, the framework adds search costs, limited awareness, spillovers or missing information, and the theory stays. This paper is a clear case: platform quality being 'inefficiently high or low' covers both directions. A framework that treats every failure as a new friction to add cannot be shown wrong by a market.
From the study
This paper develops a dynamic model of the attention economy featuring two interlinked markets: a product market, where consumers have limited awareness of firms, and an attention market, where platforms attract attention by investing in free services
The effects of data and interoperability policies can reverse over time, platform quality may be inefficiently high or low, and the rise of data-rich digital advertising can coincide with a declining ad revenue-to-GDP ratio.