[Paper Review] Macroeconomics and FinTech: Uncovering Latent Macroeconomic Effects on Peer-to-Peer Lending

Jessica Foo, Lek-Heng Lim+1 more|arXiv (Cornell University)|Oct 30, 2017

FinTech, Crowdfunding, Digital Finance6 citations

TL;DR

This paper identifies three latent macroeconomic factors—macro default, investor uncertainty, and market fundamental value—that explain systematic variation in peer-to-peer (P2P) lending interest rates across loan grade types using Canonical Correlation Analysis (CCA). In contrast, P2P credit spreads across term types show no significant macroeconomic dependence, indicating a lack of term structure in P2P lending driven by macro conditions.

ABSTRACT

Peer-to-peer (P2P) lending is a fast growing financial technology (FinTech) trend that is displacing traditional retail banking. Studies on P2P lending have focused on predicting individual interest rates or default probabilities. However, the relationship between aggregated P2P interest rates and the general economy will be of interest to investors and borrowers as the P2P credit market matures. We show that the variation in P2P interest rates across grade types are determined by three macroeconomic latent factors formed by Canonical Correlation Analysis (CCA) - macro default, investor uncertainty, and the fundamental value of the market. However, the variation in P2P interest rates across term types cannot be explained by the general economy.

Motivation & Objective

• To investigate the relationship between P2P lending rates and broader macroeconomic conditions as the FinTech market matures.
• To identify latent macroeconomic factors influencing P2P credit spreads beyond individual borrower risk.
• To determine whether macroeconomic factors explain variation in P2P interest rates across different loan grades and maturities.
• To assess the extent to which P2P lending behaves like traditional credit markets in its sensitivity to macroeconomic shocks.

Proposed method

• Employed Canonical Correlation Analysis (CCA) to extract latent factors from the joint distribution of P2P credit spreads and a set of macroeconomic variables.
• Selected macroeconomic proxies including CPI, unemployment rate (UNRATE), household debt, GDP, S&P 500 (SPX), 10-year Treasury yield, risk-free yield slope, and VIX volatility.
• Used CCA to identify three dominant canonical variates explaining systematic variation in P2P credit spreads across loan grades.
• Applied factor regression models to test the predictive power of the extracted factors on P2P credit spreads.
• Incorporated the first principal component of residuals to assess residual systematic risk not captured by the main factors.
• Conducted Wilk’s Lambda tests and redundancy analysis to evaluate the significance and explanatory power of the canonical factors.

Experimental results

Research questions

• RQ1 To what extent do macroeconomic factors explain the variation in P2P credit spreads across different loan grade types?
• RQ2 Are there latent macroeconomic factors that systematically influence P2P interest rates beyond individual credit risk?
• RQ3 Does the term structure of P2P lending (36- vs. 60-month loans) exhibit sensitivity to macroeconomic conditions similar to traditional bond markets?
• RQ4 How do investor sentiment and market uncertainty affect P2P lending spreads across different borrower risk categories?
• RQ5 Can canonical correlation analysis effectively uncover hidden macroeconomic drivers in FinTech credit markets?

Key findings

• The first canonical factor, interpreted as a macro default factor, is strongly correlated with economic expansion indicators such as rising inflation, falling unemployment, and declining household debt.
• An increase in the macro default factor is associated with higher P2P credit spreads across all grade types, indicating a risk premium during economic expansion.
• The second factor, interpreted as investor uncertainty, is positively correlated with equity volatility and the risk-free yield curve slope, and is associated with lower P2P credit spreads across grade types.
• The third factor, interpreted as the fundamental value of the P2P market, shows weak cross-loadings and limited explanatory power, suggesting limited macroeconomic pricing of market fundamentals.
• No significant macroeconomic influence was found on P2P credit spreads across term types, indicating a lack of term structure in P2P lending driven by macro conditions.
• The first principal component of residuals in the term-type regressions explains a substantial portion of remaining variation, suggesting residual systematic risk not captured by the main factors.