Revealed Wisdom of the Crowd: How Investor Bids Predict Loan Quality

Why It Matters

Online crowdfunding platforms play a crucial role in financial inclusion by connecting borrowers with investors. However, identifying high-quality loans remains a challenge. This research uncovers how the bidding behaviour of investors can reveal loan quality, helping platforms improve market efficiency and risk assessment. 

Key Takeaways  

• Investor bids contain predictive information — loans with larger bids placed early in the funding process are less likely to default.

• Crowdfunding platforms can leverage bid data — new data-driven approaches can enhance credit evaluation without compromising investor privacy. 

• Wisdom of the crowd can be better extracted — aggregating signals from informed members help isolate the truly wise part of a crowd.
  

Decoding Crowd Intelligence: Separating Signal from Noise

Despite the popularity of the phrase “wisdom of the crowd,” not all crowds are wise because not everyone in them acts in an informed, rational manner. Identifying informative actions, therefore, can help to isolate the truly wise part of a crowd. However, the challenge clearly lies in the fact that the smart crowd members are not easy to identify. They almost never have conspicuous labels attached to them, and their decision processes or judgments are rarely known.

This study leverages investors’ actions to infer their judgment and identify the more knowledgeable subset of the crowd. Unlike static demographic indicators or experience-based assessments, this approach captures real-time, behaviour-based signals that offer a more precise and timely reflection of investor wisdom. By analysing the heterogeneity in bidding behaviours, this research provides a novel way to extract meaningful insights from crowd-based investment decisions.

Investor Bidding Patterns Predict Loan Quality

Traditional credit assessment methods often struggle to capture the nuanced risk of crowdfunding loans. This study analyses investor behaviour on a major online debt crowdfunding platform, Prosper.com, to determine whether bid patterns can predict loan performance.

The study introduces the Revealed Wisdom of the Crowd (RWOC) variables, which capture investor behaviour in terms of bid size, bid timing, and deviation from typical investor actions. Using theory-driven feature engineering, the research finds that these variables significantly improve the predictive performance of state-of-the-art models that have been proposed in this context.

The Role of Early Large Bids in Loan Performance

A key insight from the study is the significance of early large bids. The paper shows that loans funded with larger bids relative to the typical bid amount in the market or to the bidder’s historical baseline, particularly early in the bidding period, are less likely to default. The rationale is that early large bids likely come from more informed investors who are confident in their assessments. Conversely, since borrower information is fixed from the start and loan values are capped, late investors who place small bids usually do so because they either lack confidence in the borrower or invest passively without scrutiny.

Business Implications

For Crowdfunding Platforms: Implementing RWOC-based predictive models can improve loan quality assessment and risk management.

For Investors: Understanding how bid timing and size reflect loan quality can guide smarter investment strategies.

For Financial Inclusion: The approach supports a more efficient and fair lending environment without relying on traditional, often exclusionary, credit scoring models.

Authors & Sources

Authors: Jiayu Yao (Nanyang Business School, NTU), Mingfeng Lin (Georgia Institute of Technology), D. J. Wu (Georgia Institute of Technology)

Article Link: https://doi.org/10.1287/mnsc.2022.02575