Consumer lending efficiency: commercial banks versus a fintech lender

Financial Innovation

Table 11 2016 relationship of financial performance measured by Tobin’s q to the ratio of nonperforming unsecured consumer loans decomposed into the Best-Practice Ratio and the Ratio in Excess of Best Practice (Lending Inefficiency)

Parameter	Variable	Coefficient estimate	Pr( >\|t\|)
β₀	Intercept	− 2.59067	< 0.0001
β₁	ln(book value of asssets_i in 1000 s)	0.42633	< 0.0001
β₂	[ln(book value of asssets_i in 1000 s)]²	− 0.01232	< 0.0001
β₃	Equity capital_i/book value of assets_i	0.32866	0.0019
β₄	Unsecured consumer loans_i/total loans_i	0.20179	0.1004
β₅	Best-practice nonperforming consumer loans_i/total unsecured consumer loans_i	1.55851	0.0647
β₆	Excess nonperforming consumer loans_i/total unsecured consumer loans_i (lending inefficiency)	0.04723	0.5153
β₇	[Unsecured consumer loans_i/total loans_i] × [best-practice nonperforming consumer loans_i/total unsecured consumer loans_i]	− 5.57497	0.0762
β₈	[Unsecured consumer loans_i/total loans_i] × [excess nonperforming consumer loans_i/total unsecured consumer loans_i (lending inefficiency)]	− 11.78893	0.0094
# of entities	∂(q ratio)/∂(best-practice np ratio) > 0	N = 202
	significantly > 0	N = 182
# of entities	∂(q ratio)/∂(best-practice np ratio) < 0	N = 3
	significantly < 0	N = 0
# of entities	∂(q ratio)/∂(inefficiency ratio) > 0	N = 33
	significantly > 0	N = 0
# of entities	∂(q ratio)/∂(inefficiency ratio) < 0	N = 172
	significantly < 0	N = 102

The data set includes 205 top-tier publicly traded bank holding companies at the end of 2016 with plausible values of nonperforming unsecured consumer loans and total loans exceeding 10 percent of assets. Financial performance is gauged by Tobin’s q ratio. Nonperforming loans are the sum of past due and nonaccruing loans and gross charge-offs. Total unsecured consumer loans include gross charge-offs. The decomposition of the nonperforming unsecured consumer loan ratio into the best-practice ratio (inherent credit risk) and the ratio in excess of best practice (lending inefficiency) is obtained from a stochastic frontier that defines a lender’s peers as those with credit risk similar to that of the lender’s consumer loan portfolio, similar economic characteristics of the lender’s local markets, such as the weighted 10-year average GDP growth rate and the weighted average Herfindahl index across these markets, where the weights are bank deposit shares, a similar 3-year growth rate of the lender’s consumer lending,and a similar volume of its consumer lending. Statistical significance is computed from robust standard errors. Bold values indicate statistical significance at stricter than 0.10. Adjusted R² = 0.6874. AIC \(=-\!\)1372