GDP-B: Accounting for the Value of New and Free Goods
Source: https://www.semanticscholar.org/paper/9c229b94e2c51c9a3d4bced7380368eb2d3abb6d ↗
Brynjolfsson and colleagues tackle the central measurement problem of the digital economy: how do you account for the welfare value of free products like search engines, social networks, or smartphone apps that never appear in GDP statistics? Their GDP-B framework uses incentive-compatible choice experiments to estimate what people would pay to keep these products, revealing billions in unmeasured value.
For product directors this is essential infrastructure — a rigorous way to think about value creation beyond revenue metrics.
The paper provides both theoretical foundation and practical methodology for measuring impact in platform economies where traditional accounting breaks down.
The smartphone camera example alone demonstrates why product teams need better frameworks for understanding the economic value they create beyond what shows up in financial statements.
Central argument
Brynjolfsson and colleagues argue that GDP systematically undercounts economic welfare in the digital age because it only captures market transactions, leaving the consumer surplus of free goods — search engines, social networks, apps — entirely invisible. To fix this, they propose GDP-B (B for 'beyond GDP'), a complementary national accounting metric derived from large-scale incentive-compatible choice experiments in which participants reveal how much they would need to be paid to give up specific digital goods for a month. Their key empirical finding is that these willingness-to-accept valuations are enormous relative to the negligible revenue these products generate, meaning conventional productivity statistics structurally understate the value created by the digital economy.
Critique
The methodology's reliance on willingness-to-accept (WTA) valuations from hypothetical deprivation experiments is vulnerable to a well-documented behavioral economics problem: WTA estimates are notoriously unstable and tend to be inflated relative to actual revealed preferences, partly because participants anchor on their attachment to a good rather than its substitutability. This means GDP-B may overstate welfare gains, especially for products with plausible alternatives that participants don't fully consider in a lab setting. More fundamentally, aggregating individual survey responses into a national statistic introduces the same index-number problems the authors critique in conventional GDP, raising questions about whether GDP-B is genuinely more rigorous or simply relocates the measurement difficulty.
Why it matters for product
For a product director, GDP-B reframes the core strategic question: if your product is free, your team's value creation is almost entirely absent from every financial metric your organization tracks, which means OKRs and business cases built on revenue or GMV are measuring the shadow of the product's actual impact. The paper's methodology — asking what users would pay to avoid losing the product — is directly actionable as a product discovery and prioritization tool, giving teams a defensible way to rank features or surface bets by welfare value rather than monetization potential alone. This is particularly sharp for platform leaders navigating the tension between growth metrics and genuine user value, where the gap between what shows up in dashboards and what users actually depend on can drive misinformed roadmap decisions.