August 10, 2015

David SilbermanAssociate DirectorDivision of Rules, Markets & ResearchConsumer Financial Protection Bureau1700 G Street, NWWashington, D.C. 20552

Re: Potential Rulemaking for Payday, Vehicle Title and Similar Loans

Dear Mr. Silberman:

On March 26, 2015, the Consumer Financial Protection Bureau (“CFPB” or “Bureau”) announced its intention to make rules regulating the market for payday, vehicle title and similar loans. The CFPB also released a detailed outline of the various alternatives that it is considering for regulating this market. As they are set forth in the outline, the CFPB’s policy proposals are “very complex, which could hinder compliance, transparency, and enforcement.”1 We are writing to you in anticipation of a formal CFPB proposal to regulate payday, vehicle title and similar loans.

I. About Online Lenders Alliance........................................................................................3

II. CFPB’s Perceived Consumer Injury...............................................................................4

III. Alternatives Under Consideration...................................................................................4

A. Covered Short-Term Loans..................................................................................4

B. Covered Longer-Term Loans...............................................................................5

C. Requirements Applicable to All Covered Loans................................................6

IV. Ability to Repay (Residual Income Test).........................................................................7

A. The Proposal Ignores Sound Underwriting To Focus On Residual Income.....................................................................................................................7

B. The CFPB Has Considered, and Sidestepped, a Residual Income Test in Other Contexts...................................................................................................8

1 Pew Charitable Trusts, Understanding the CFPB Proposal for Payday and Other Small Loans (July 6, 2015).

Online Lenders Alliance | P.O. Box 320130 | Alexandria, VA 22320(703) 567-0327 | www.oladc.org

C. The Residual Income Test Burdens Consumers and Lenders, and Needlessly Reduces Access To Credit................................................................10

D. Underwriting Must be Flexible..........................................................................11

E. Residual Income Unnecessarily Restricts Credit, Without Measuring Affordability.........................................................................................................11

F. The Residual Income Driven Ability to Repay Determination Disadvantages Small Lenders.............................................................................12

G. Residual Income Test is Not Consistent with Regulation B.............................12

H. Alternatives to Residual Income.........................................................................12

V. Collection and Verification.............................................................................................13

A. The CFPB’s Proposal Diminishes Competition................................................13

B. Collection and Verification of Documents Disadvantages Small Lenders.................................................................................................................14

C. The CFPB’s Proposal for Document Collection, Verification and Retention Is Ineffective And Burdensome.........................................................14

D. Document Collection And Verification Requirements Present A Burden For Consumers And Frustrate Consumer Choice..............................15

E. The Proposal Raises Significant Data Privacy And Security Concerns.........16

F. The Requirement To Collect and Verify Documents Should Permit Innovation By Online Lenders...........................................................................17

VI. Payment Authorization Does Not Expose Consumers to Additional Risk.................18

A. Consumers Are in Control of Their Own Accounts.........................................18

B. Providers Should Not Be Discouraged From Offering The Convenience Of Recurring Payments To Consumers......................................19

C. Consumer-Authorized Debits Do Not Receive Preferential Treatment.........19

D. Effect of Overdraft Protection............................................................................20

E. Timing Of Payments Protects Consumers........................................................21

F. New NACHA Rules Will Change The Payments Landscape..........................21

2

G. Debit Cards Payments Are Different Than ACH Payments and Should Not be Subject to the Covered Loan Requirements............................22

VII. Loan Limits Do Not Protect Consumers.......................................................................22

A. Flexibility in Loan Limits....................................................................................23

VIII. The Federal Database Proposal Is Not Workable........................................................24

IX. The Evidence Shows that the Market for Short-Term Loans Works........................25

X. The 36% Military APR Trigger For Rule Coverage Is Not in the Public Interest..............................................................................................................................26

A. APR Is a Flawed Measurement, and Military APR Is Doubly Flawed..........26

B. Small Dollar Loans Cannot Be Made Economically at 36% APR..................28

C. The CFPB’s Rate Cap Will Hurt Consumers...................................................30

XI. The Proposal Would Reduce Competition, Innovation, Consumer Choice, and Access to Credit........................................................................................................30

A. The Proposal Could Cause A Migration To Inferior Alternatives.................31

B. Innovators Are Often Higher Cost Providers...................................................32

XII. Alternative Solutions.......................................................................................................32

I. About Online Lenders Alliance

The Online Lenders Alliance (“OLA”) represents the growing industry of companies offering online consumer loans and related products and services. OLA members include online lenders, and vendors and service providers to lenders, such as consumer reporting agencies. Besides complying with applicable federal consumer financial protection laws, OLA member companies have agreed to comply with additional requirements under the organization’s Best Practices and Code of Conduct, which offer additional protections for consumers. OLA supports efforts by the CFPB, Federal Trade Commission, federal banking agencies and others to halt bad actors who engage in deceptive, unfair or abusive lending practices.

OLA has a particular interest in the CFPB’s alternatives under consideration, because the online lending market will be uniquely and adversely impacted by the deeply flawed standards outlined in these proposals. The increased regulation proposed for longer-term loans would be triggered by a borrower’s choice of electronic payment, and online lenders rely on electronic means both to fund and service the loans that they extend to consumers. Many other lenders regularly extend

3

credit in excess of 36% APR, but because they are not repaid electronically, the alternatives under consideration would not apply to them.

II. CFPB’s Perceived Consumer Injury

In support of its alternatives under consideration, the CFPB has stated that single payment loans due within one pay cycle create a “debt trap” for consumers when they are unable to repay the loan without re-borrowing.

The CFPB also asserts that account or payroll access or a security interest in a vehicle discourages lenders from adequately underwriting credit.2 According to the CFPB:

The practice of a borrower authorizing an account debit for repayment of a loan can “result in many consumers taking on unaffordable loans,” and can force the borrower to “default on other obligations or to reborrow or refinance the loan.”

Borrowers “may not know when presentments will be made, in what amount, and for what reason. As a result, consumers may be unable to move money into the account to cover the presentment or, alternatively, to stop payment on the presentment.”

“[S]ome lenders continue to present items for payment after multiple prior failed attempts. If a consumer’s account lacks sufficient funds to cover a payment when the lender seeks to collect payment from the consumer’s account, the consumer may incur substantial costs.”

III. Alternatives Under Consideration

In its outline of the various alternatives under consideration, the CFPB sets out two separate regulatory schemes – the first for “covered short-term” loans (those under 45 days) and the second for “covered longer-term” loans. Covered longer-term loans would be loans with a term longer than 45 days, with an “all-in” APR in excess of 36%, where the borrower has authorized the lender to access the borrower’s checking account, payroll or automobile title for repayment.

A. Covered Short-Term Loans

For short-term loans, lenders would be required to make a reasonable determination that a loan applicant can repay the loan when due—including interest, principal and fees for add-on products—without reborrowing within underwriting period, which is defined as the loan term plus 60 days.

For each loan, lenders would be required to verify the amount and timing of the applicant’s income and the amount and timing of major financial obligations to make a reasonable determination whether the applicant would have enough residual income left to repay the loan

2 See Consumer Financial Protection Bureau, Outline of Proposals Under Consideration and Alternatives Considered, at 19-22, 29 (Mar. 26, 2015).

4

after covering other major financial obligations and living expenses. In addition, the applicant could not have an outstanding covered loan with any lender, and the lender would be required to check the applicant’s credit history with at least one credit bureau approved by CFPB for this purpose. The CFPB also indicates that the lender should consider past defaults and re-borrowing by the applicant, although the alternatives under consideration do not appear to prohibit a lender from extending credit to an applicant with a history of defaults or re-borrowing.

The CFPB would propose to limit re-borrowing by imposing a 60-day cooling-off period between covered short-term loans, unless lender can document an applicant’s improved circumstances (such as, additional income) with “verified evidence.” In addition, the CFPB would prohibit a consumer from borrowing more than three loans in a “sequence,” defined as a series of loans for each loan is extended within 60 days of repayment of the prior loan.

CFPB would offer a streamlined compliance option, whereby a lender would not be required to make an ability to repay determination, but would still be required to verify income and borrowing history and also to verify that the applicant does not have an outstanding covered loan with any lender. Under the streamlined compliance option, a lender could make three consecutive loans, after which a mandatory 60-day cooling-off period would be imposed. The second and third consecutive loans would be permitted only if the lender offers an affordable way out of debt – referred to by the Bureau as an “off-ramp.” The CFPB is considering two options for the off-ramp: (1) requiring that the principal decrease with each loan so that it is repaid after the third loan; or (2) requiring that the lender provide a no-cost payment plan for the third loan to allow the consumer to pay the loan off over time without further fees. For each loan meeting these requirements, the debt could not:

exceed $500; carry more than one finance charge; require the consumer’s vehicle as collateral; result in the consumer being indebted for more than 90 days in any rolling 12 month

period; or result in a consumer obtaining more than six covered short-term loans in any rolling 12

month period.

B. Covered Longer-Term Loans

With respect to “covered longer-term loans,” the CFPB would require lenders to make a reasonable determination that a loan applicant can repay the loan when due—including interest, principal and fees for add-on products—without reborrowing during the loan term. (For a longer-term loan with a balloon payment, lenders would be required to ensure that an applicant could repay the loan without re-borrowing during the loan term plus 60 days.)

As with covered short-term loans, lenders would be required to verify the amount and timing of the applicant’s income and the amount and timing of major financial obligations to make a reasonable determination whether the applicant would have enough residual income left to repay the loan after covering other major financial obligations and living expenses. In addition, the applicant could not have an outstanding covered loan with any lender, and the lender would be

5

required to check the applicant’s credit history with at least one credit bureau approved by CFPB for this purpose.

For longer-term loans without balloon payments, there is no limit on renewals, but there is a presumption that an applicant would be unable to repay a loan if the applicant is seeking to refinance any loan with the lender, or a covered loan with another lender, and (1) the loan is delinquent or in default, or (2) the consumer “indicates” that he or she cannot make payments or the loan is causing distress, or (3) the refinancing allows the consumer to skip payment or reduce his or her payment. As with short-term loans, a lender can rebut this presumption by documenting a change in circumstances.

For longer-term loans with a balloon payment, there is a presumption that there is no ability to repay if the consumer attempts to borrow during the term of, or within 60 days after, a covered loan with a balloon payment outstanding. A lender can rebut this presumption by documenting a change in circumstances. Following three covered longer-term loans with a balloon payment (or short-term loans, or a combination) in a sequence, the presumption of inability to repay is conclusive, and a 60 day cooling-off period is imposed.

The CFPB sets out two potential streamlined compliance options for lenders extending covered longer-term loans. For both of these options, the loan must be fully amortized over no fewer than two payments; the lender must verify the consumer’s income and borrowing history; the lender must report the loan to all applicable commercially available reporting systems; and loan terms cannot be longer than 6 months.

NCUA Loan Option: Lenders may provide generally the same protections offered under the National Credit Union Administration (NCUA) program for “payday alternative loans.” These loans have a 28% interest rate cap and an application fee that reflects the actual costs of processing the application (not to exceed $20).

Five Percent GMI Option: The second option would allow lenders to make a longer-term loan if the amount the consumer is required to repay each month is no more than 5% of the consumer’s gross monthly income and there is no fee for prepayment. The lender could not make more than two of these loans within a 12-month period.

C. Requirements Applicable to All Covered Loans

The Bureau would also impose requirements applicable to both short-term and longer-term lenders. Lenders of covered loans would be required to provide borrowers with a reminder that payment is due three business days prior to the due date. In addition, if two consecutive attempts to collect money from the borrower’s account are unsuccessful, the lender would be required to obtain a new payment authorization from the borrower.

Lenders of covered loans would be required to report borrower’s loan performance to all credit bureaus approved by CFPB for this purpose, and also would be required to check the credit history of all applicants for covered loans with at least one credit bureau approved by CFPB for this purpose.

6

Lenders of covered loans also would be required to establish and maintain written policies and procedures for compliance with the CFPB rules, such as written policies and procedures for determining ability to repay.

Lenders of covered loans also would be required to keep records documenting actions taken with respect to a covered loan until 36 months after the last entry on the loan. These records would include documentation of a consumer’s income, major expenses and borrowing history.

IV. Ability to Repay (Residual Income Test)

The CFPB’s proposal would require small dollar lenders to make a “reasonable determination” that an applicant will be able to repay the loan. To make this determination, a lender would need to determine that the consumer will have enough residual income to repay the covered loan, fulfill his or her major financial obligations and meet living expenses without re-borrowing during the underwriting period. The lender would need to verify these facts by reviewing the consumer’s pay stubs, bank statements, lease, canceled checks, and credit reports.

The CFPB’s test stifles innovation, mandates the use of untested residual income underwriting standards, and sets in stone antiquated, unsafe, and unreliable methods of data collection and verification. Such an approach is hardly what regulated entities or the general public should expect from a regulator that prides itself on being a forward-looking, 21st century agency. Online lenders have been responsible for many of the recent innovations in online payments, fraud detection and prevention, credit risk underwriting, and account servicing. Online lenders have also been leaders in developing innovative underwriting tools and algorithms to predict with a great deal of accuracy whether a consumer is likely to repay a loan.3 Modern underwriting relies on algorithms, not paperwork. Even as lenders devise new and better ways of determining consumers’ creditworthiness, they will still need to build a budget and determine residual income to comply with the CFPB’s rule.

A. The Proposal Ignores Sound Underwriting To Focus On Residual Income

While an ability to repay determination seems reasonable, it is only one part of a creditworthiness determination. The proposal overlooks an equally important factor that demonstrates a borrower’s creditworthiness: willingness to repay. Lenders generally use a borrower’s history of repayment, expressed through credit scores and other analytics, as an indicator of willingness to repay. A borrower’s residual income, or capacity to repay, does not mean that he or she will actually repay.

3 See, e.g., Steve Lohr, “Banking Startups Adopt New Tools For Lending, ” New York Times, Jan. 15, 2015; Colin Wilhelm, “Are Online Lending Platforms Beating Banks and Big Data?,” American Banker, November 5, 2014 (subscription required); Kevin Wack, “Maximizing Big Data to Minimize Risk in Small-Dollar Lending,” American Banker, July 22, 2013 (subscription required). See also Pew Charitable Trusts, Harmful Practices in Internet Payday Lending, at 4 (“Many online lenders use sophisticated technology and advanced algorithms to predict which applicants are most likely to repay loans.”).

7

Furthermore, the proposal would allow the CFPB to potentially designate the lender’s “ability to repay” methodology as unreasonable if the lender’s borrowers consistently refinanced or defaulted on their loans. This ignores the fact that refinancing and defaults may indicate a borrower’s lack of willingness to repay, or a borrower’s particular credit needs at that time, not a lack of ability to repay.

A statistical analysis of records of 87 million payday loans made by five large lenders in 37 states over more than five years, found that a higher payment-to-paycheck ratio of the borrower (which indicates a greater ability to repay) does not correlate with a lower probability of default.4 The CFPB has not addressed this important finding, nor has it otherwise established that residual income is determinative of creditworthiness.

B. The CFPB Has Considered, and Sidestepped, a Residual Income Test in Other Contexts

The CFPB considered a residual income test, along with a debt-to-income ratio (DTI) test, in the ability to repay/qualified mortgage (ATR/QM) rule.5 In that rule, the CFPB provided detailed guidance in Appendix Q on how to calculate a DTI ratio and used the DTI ratio as a key factor for satisfying the qualified mortgage safe harbor and rebuttable presumption of compliance.

Even though the CFPB retained the concept of residual income in the rule, it did not find it “necessary or appropriate to specify a detailed methodology in the final rule for consideration of residual income.”6 Instead, the CFPB adopted “broad standards for the definition and calculation of residual income” to preserve creditor flexibility to develop and refine more nuanced residual income standards in the future.7 These “broad standards” require creditors to make “reasonable and good faith determinations of the consumer’s repayment ability in light of the facts and circumstances.”8 The only guidance the CFPB provided on residual income was to suggest that a consumer’s residual income, if high enough, may compensate for a higher DTI ratio in an ability to repay determination.9

The CFPB recognized that few lenders use residual income in underwriting: “Except for one small creditor and the VA, the Bureau is not aware of any creditors that routinely use residual income in underwriting, other than as a compensating factor.”10

4 See Peter Toth, Measures of Reduced Form Relationship Between the Payment-Income Ratio and the Default Probability (Feb. 27, 2015).5 78 Fed. Reg. 6,408 (Jan. 30, 2013). 6 Id. at 6,487.7 Id. 8 Id. at 6,486.9 Id. 10 Id.

8

As for the VA residual income test, the CFPB found that it did not have sufficient information to evaluate whether the VA’s approach is truly predictive of repayment ability:

The VA underwrites its loans to veterans based on a residual income table developed in 1997. The Bureau understands that the table shows the residual income desired for the consumer based on the loan amount, region of the country, and family size, but does not account for differences in housing or living costs within regions (for instance rural Vermont versus New York City). The Bureau also understands that the residual income is calculated by deducting obligations, including Federal and State taxes, from effective income. However, at this time, the Bureau is unable to conduct a detailed review of the VA residual income guidelines, which would include an analysis of whether those guidelines are predictive of repayment ability, to determine if those standards should be incorporated, in whole or in part, into the ability-to-repay analysis that applies to the entire residential mortgage market.11

Based on this record, the CFPB essentially acknowledged that it did not know how to construct a residual income calculation. The CFPB indicated that it would “consider conducting a future study on the debt-to-income ratio and residual income[,]” but such a study has not, to our knowledge, been undertaken by the Bureau.12 Despite these facts, the CFPB now proposes using an untested and undefined residual income test for making ability to repay determinations for small-dollar loans.

Given this history, the CFPB’s proposal to adopt a residual income test as the sole test for determining a borrower’s ability to repay is undermined by:

Its previous acknowledgement that residual income is not routinely used in credit underwriting, other than as a compensating factor, such as to override a high DTI ratio;

Its previous acknowledgement that it has no evidence that the VA residual income standards, the only such standards known to exist, are predictive of repayment ability;

Its failure to conduct and publish a detailed review of the VA residual income standards to determine if those standards are an appropriate test for determining a borrower’s ability to repay; and

Its failure to specify how lenders should account for variables, such as local differences in housing and living costs.

11 Id. at 6,486-87.12 Id. at 6,486.

9

C. The Residual Income Test Burdens Consumers and Lenders, and Needlessly Reduces Access To Credit

The residual income test imposes a difficult and confusing budgeting task on borrowers and lenders. The CFPB’s test requires borrowers and lenders to undertake the arduous task of developing a household budget, and then determining how much residual income is left to make debt service payments. Borrowers and lenders are not even be required to do this for a $500,000 mortgage, where the CFPB’s Qualified Mortgage rules provide alternatives to residual income determinations.

A more practical approach would be based on the credit card ability to repay provisions, where credit card issuers make ability to repay determinations based on self-reported income and debts reflected in a consumer report. See 12 C.F.R. § 1026.51. To complete this analysis, card issuers generally use the income that is provided by the consumer on his or her card application, which is expressly permitted under the commentary to Regulation Z. See 12 C.F.R. § 1026.51(a)(1)(i)-5.i. Credit card issuers generally obtain information relating to obligations by reviewing the trade lines set forth on a consumer’s credit report, which is also permitted under the commentary to the regulation. See 12 C.F.R. § 1026.51(a)(1)(i)-7. If this analysis is sufficient for credit card loans, including subprime loans, with much higher balances than the typical payday or title loan, it should also be acceptable for small dollar loans. There is no rational basis for the CFPB to impose a more burdensome ability to repay requirement on a non-bank lender making a short-term small dollar loan than a large bank making a potentially much larger subprime credit card loan.

Furthermore, the CFPB’s standard would limit access to credit by creditworthy borrowers. If, as shown, residual income has little or no bearing on creditworthiness, lenders will be forced to deny creditworthy applicants who do not pass the CFPB’s novel and untested residual income test. The ability to repay determination would exclude consumers who would otherwise have received and successfully repaid the loan.

Residual income imposes an impossible standard – lenders have to underwrite each payment, including timing and amount of income, and timing and amount of other obligations. For example, what happens in the two-week period when the rent comes due? In addition, what happens when a consumer’s income is sporadic, seasonal, or temporarily interrupted? Under these circumstances, the residual income test risks unnecessarily depriving these consumers of credit. The CFPB made certain accommodations for these types of circumstances in the ATR/QM rules,13 but, despite these adjustments, even CFPB Director Cordray recently recognized the Bureau has continuing concerns about lack of access to mortgage credit for self-employed, temporary, and seasonal workers.14 These populations with erratic income patterns 13 See 12 C.F.R. Part 1026, Appendix Q (containing special rules for calculating DTI ratios with respect to income for seasonal employment, self-employed consumers, and overtime and bonus income). 14 The Consumer Financial Protection Bureau’s Semi-Annual Report to Congress: Hearing Before the Comm. on Banking, Housing, and Urban Affairs, 114th Cong. (July 15, 2015) (testimony of R. Cordray).

10

have the greatest need for the flexible, unsecured credit products offered by online lenders to help “smooth” their income. Yet, the CFPB’s residual income test would effectively deprive these borrowers of access to such credit.

D. Underwriting Must be Flexible

Prescribed limits – such as 5% GMI – will not work because they are static measure for underwriting. In the business world, lenders work with constantly changing data in developing customer relationships for effective underwriting. Underwriting models are developed and constantly refined by individual lenders through their experience with their own portfolios and customer populations. Constraining underwriting to a specific formula will prevent utilization of useful data and will stifle innovation.

In addition, an inflexible mandatory residual income-based ability to repay model does not permit lenders to make special rules for special categories of borrowers, and does not allow lenders to make discretionary exceptions where the lender believes a risk is acceptable.

E. Residual Income Unnecessarily Restricts Credit, Without Measuring Affordability

The reason consumers borrow $500 over a relatively short period of time is that they don’t have enough residual income to make ends meet – perhaps an emergency or illness has resulted in a temporary shortfall in income or an increase in expenses. Even if consumers do not have enough income on paper, most still manage to repay these loans – for example, by prioritizing their other payments and obligations until a temporary shortfall in income or increase in expenses passes. The proposed ability to repay standard is too narrow. It focuses only on whether there is enough residual income to make the payment, and does not consider other means by which consumers successfully repay – such as by prioritizing other obligations. For example, a consumer may understand that a utility bill need not be paid immediately, but could be paid on the second or third notice without consequence. The residual income test also does not consider consumers who might anticipate and plan for renewal of the loan.

Many consumers experience expected or unexpected fluctuations in income, debts, or both for a variety of reasons. These loans are frequently used to get through those rough patches in a person’s financial life. The residual income test makes it impossible to access this type of credit in the very circumstances where it is most needed by consumers.

F. The Residual Income Driven Ability to Repay Determination Disadvantages Small Lenders

Requiring unproven and novel underwriting standards, coupled with costly, unsafe, and antiquated documentation requirements, threaten to disrupt the viability of the entire small dollar lending market, but would disproportionately impact startups and small businesses, which likely could not remain in this market. Whether the ability to repay determination is manual or automated, it is in addition to a lender’s existing underwriting processes. Each lender will need to build that a new platform to make ability to repay determinations that are consistent with the

11

CFPB’s expectations. The costs of building this platform will be much lower on a per unit basis for larger lenders than they will be for smaller lenders.

G. Residual Income Test is Not Consistent with Regulation B

Lenders would need to ask sensitive and possibly unlawful questions about applicants’ marital status and information concerning a spouse to properly apply the residual income test and understand the effect of applicants’ income and household expenses.

Regulation B prohibits lenders from asking for information about spouses, unless an applicant is relying on a spouse’s income for repayment. The CFPB must consider how lenders can obtain information about household expenses and calculate a borrower’s residual income without violating this important and longstanding regulatory requirement. Similarly, the CFPB must consider how lenders should apportion household expenses and obligations between an applicant’s income and a spouse’s income to calculate an applicant’s residual income without violating Regulation B.

H. Alternatives to Residual Income

The CFPB’s goals surely could be achieved by less costly means. The CFPB may be trying to ensure that fewer consumers default on small dollar loans, or trying to ensure that borrowers are not repaying small dollar loans by overdrawing their deposit account. If so, then the CFPB should consider using its existing supervisory authority in this market15 to determine the different methods, models and algorithms that lenders use to manage credit risk, the efficacy of those credit risk management tools, and require that lenders periodically revalidate models to determine that they are performing as intended.16 Where a lender is not achieving what the CFPB considers to be an acceptable rate of default, it can be addressed as a supervisory matter, rather than trying to fit the square peg of underwriting policies and procedures into the round hole of unfair and abusive practices.

An informational remedy would be more effective and less restrictive. If the CFPB is concerned about borrowers’ understanding of the all-in costs of a loan or the remedies available to a lender, the solution is more and better disclosure, not a rule that in effect eliminates borrowers’ ability to access the product. As President Obama has written, “giving careful consideration to benefits and costs . . . means using disclosure as a tool to inform consumers of their choices, rather than restricting those choices.”17 15 The Dodd-Frank Act authorizes the CFPB to supervise “payday lenders.” 12 U.S.C. § 5514. The CFPB has defined its jurisdiction very broadly, stating that “payday loans” are “small-dollar; borrowers must repay loan proceeds quickly (i.e., they are short-term); and they require that a borrower give lenders access to repayment through a claim on the borrower’s deposit account.” CFPB, Short-Term, Small Dollar Lending Examination Procedures at 2.16 The other bank regulators have long imposed similar requirements on banks. See, e.g., OCC 2011-12, Supervisory Guidance on Model Risk Management (April 4, 2011).17 Barack Obama, Toward a 21st-Century Regulatory System, Wall Street Journal (Jan. 18, 2011).

12

In addition, the CFPB should create an exemption from the ability to repay requirement for any lender with below-market-average charge-off rates. This would create an ongoing incentive to encourage lenders to continuously improve their underwriting models and offer loans that can and will be repaid. As average charge-off rates drop, lenders seeking to avoid the burdens of the ability to repay analysis would be forced to fine tune their underwriting criteria to remain below the average.

V. Collection and Verification

The CFPB’s proposal to regulate small dollar loans would require lenders to collect and verify an immense amount of information about consumers’ incomes, household expenses and credit history, and retain such information for three years, longer than record retention requirements under existing consumer protection law.

A. The CFPB’s Proposal Diminishes Competition

Online lenders will need to invest in new systems to collect, verify and retain documents, assuming that it is even possible to collect and verify paper documentation in an online environment. In addition to these fixed costs, lenders will incur higher incremental costs for the labor and credit reporting involved.

In 2010, the State of Washington amended its Check Cashers and Sellers statute.18 Changes included a paystub/income verification requirement and a 30% Gross Monthly Income requirement. The chart below shows a substantial decrease in the number of licensed payday lender locations (71%) and the number of loans made (73%) in the State of Washington since 2009:

2006 2007 2008 2009 2010 2011 2012 2013

Companies 130 138 133 109 85 68 52 46

Branches 612 591 584 494 339 188 151 128

Total Locations 742 729 717 603 424 256 203 174

Loans Made (000’s) 3504 3266 3197 3244 1094 856 910 872

Source: Washington State Department of Financial Institutions, 2013 Payday Lending Report.

The CFPB’s rule will have an even more dramatic effect on the market. Lenders who survive the CFPB’s rule are likely to raise prices, both to take advantage of their larger market share and to recoup the foregone revenue and increased expenses associated with complying with the rule. Ultimately, consumers will be harmed.

18 RCW §§ 31.45.010 et seq.

13

B. Collection and Verification of Documents Disadvantages Small Lenders

Developing the policies, procedures, systems, employees, call centers and other facilities for collecting and verifying documentation relating to income, major obligations, and borrowing history imposes a disproportionate burden on small businesses, putting those businesses at a competitive disadvantage.

Large businesses will have the capital and customer base to make the needed upfront investment in the systems and facilities. In addition, larger businesses will be able to spread the cost of compliance over a larger number of loans, for a lower per unit cost. Small lenders, unlike their large competitors, will not achieve the volume to offset their high fixed costs, nor the economies of scale to offset their high incremental costs of compliance.

C. The CFPB’s Proposal for Document Collection, Verification and Retention Is Ineffective And Burdensome

The CFPB’s antiquated proposals as to the form, manner, and amount of required information collection and verification about income, obligations and borrowing history constitute a deeply flawed and profoundly risky approach to attempting to predict ability to repay that imposes unnecessary burdens on consumers and businesses. Because these documents do not improve underwriting, the collection and verification of them amounts to needless red tape and expense.

Lenders would be required to verify the amount and timing of a consumer’s income through bank statements, benefit statements, or paystubs. Documentation of income (pay stubs) and obligations (bank statements), however, is highly susceptible to fraud. For those consumers who are intent on defrauding lenders, pay stubs and bank statements or easily doctored or fabricated.19

Documents submitted by fax, mobile image capture or scanned and emailed are frequently not legible or easily interpreted. Moreover, mobile image capture does not work for pay stubs and other irregular documents. In order for a digital image capture to be effective, the documents being captured must be in a standard format for companies to process the image automatically and correctly. That is, the software can read the appropriate information in order to process the information only when the information is in the same format and in the same location – for every single document. This works with checks deposited to bank accounts because the information read by the software is (1) all numerical and (2) located in the same location on the check. By contrast, pay stubs and other documents the CFPB would require lenders to collect are not standardized. Even if customers could submit these documents electronically via a mobile app, the documents would still need to be manually processed and verified on the back end by lenders.

D. Document Collection And Verification Requirements Present A Burden For Consumers And Frustrate Consumer Choice

19 See http://www.paystubb.com/ and numerous similar websites.

14

Consumers without access to a scanner, fax machine or recent-model smartphone would have significant difficulties providing documentation to the lender that supports their ability to repay. The CFPB should not erect a technological barrier to credit for consumers.

In addition, some borrowers work cash-only jobs where no pay stub is readily accessible, or freelance, seasonal, or hourly jobs where pay stubs may not be representative of their short-term future earnings. Other borrowers – such as an individual living with a family member – may be unable to document expenses. These borrowers may be creditworthy but cannot document their ability to repay to the standards of the proposal, and the CFPB’s requirements will shut these creditworthy borrowers out of the system. This issue also has been raised with respect to the ability to repay requirements for mortgages, and in recent testimony, Director Cordray said that the CFPB intends to revisit the income-verification standards under the Ability to Repay Rule. According to Director Cordray, the CFPB is “increasingly aware of and concerned about” the possibility that self-employed, temporary and seasonal workers may see limited mortgage credit due to documentation requirements. “It’s not an easy thing to figure out how to handle, but it's something we need to go back and think more about.”20

We also understand that many states – Alabama, Arkansas, Florida, Louisiana, Mississippi, Nebraska, Ohio, South Dakota, and Tennessee – do not require employers to give statements of wages or earnings to employees. Other states require that a statement contain deductions to the employee’s wages or earnings, but they do not require that employers state the actual wages or earnings on the statement. Still other states, such as Georgia and Ohio, require employers to give paystubs only in certain circumstances. Consumers in these states have no way to compel their employers to produce a pay stub.

Furthermore, consumers in the online credit market value speed and convenience,21 and the CFPB’s proposed requirements will frustrate consumers’ expectations. The time required to obtain and verify detailed documentation regarding income and obligations will drive consumers to inferior alternatives, such as overdraft protection or nonpayment of utilities, resulting in reconnection fees.

E. The Proposal Raises Significant Data Privacy And Security Concerns

The proposal would require lenders to collect information about consumers’ income, assets, debts, and borrowing history, and to keep such information for 36 months after the last entry on the loan. The proposal is inconsistent with FTC data security guidance and rules, such as the GLBA Safeguards Rule, which states that companies should collect the minimum amount of information necessary, and keep that information for as little time as possible. 20 The Consumer Financial Protection Bureau’s Semi-Annual Report to Congress: Hearing Before the Comm. on Banking, Housing, and Urban Affairs, 114th Cong. (July 15, 2015) (testimony of R. Cordray).21 FDIC, National Survey of Unbanked and Underbanked Households, at 41 (Dec. 2009); Texas Appleseed, Short-Term Cash, Long-Term Debt: A Survey of Payday Borrowers, at 9 (Apr. 2009); American Banker, Five Reasons You Can’t Ignore the Neobanks (Apr. 27, 2015); Pew Charitable Trusts, How Borrowers Choose and Repay Payday Loans, at 42 (Feb. 2013).

15

Moreover, the CFPB encourages consumers to transmit sensitive documents by taking photos with their smartphones, which is not a secure transmission method. Photos and documents sent via email are not encrypted and can be intercepted by third parties – at any point during their transmission. Such emails also leave a trail on every single mail server that they traverse to get to the recipient from sender. For example, if a customer at a coffee shop decided to email a photo of a drivers’ license, anyone who is also on the coffee shop network could potentially access the email and attachment. Even if a customer sends an email while on his/her secure home network, the email and attachment could be visible to anyone (and will leave the same trail) once it leaves the home network. There is no way for a customer or a recipient to know whether the email and attachment was monitored or intercepted, thus customers cannot know whether they need to take action to protect their identity.

The CFPB proposal would replace a system that is secure for both consumers and lenders and replace it with an antediluvian system of paper-based or insecure electronic document and collection and verification techniques that presents heightened security risks to both parties, an anachronistic approach that flies in the face of current government and industry efforts to enhance data security and privacy in the modern digital world. Today, our members’ customers typically enter information on a web app that is secured by https and obfuscated so that when customers type in information, the characters are immediately hidden. Information transmitted through the app is encrypted, unlike when a customer emails documentation. Lenders can proactively monitor and scan their websites and systems for data breaches by utilizing multiple separate independent vendors at various security layers. Each vendor performs a set of comprehensive application and network penetration tests or assessments based on regulatory standards and best practices. Lenders’ websites can be protected by a zone-based architecture with layered security components including routers, firewalls, network intrusion detection systems and host intrusion detection systems. Lenders also employ web application code scanning to detect any security defects before any application is released into production.

Under the CFPB proposal, instead of just entering information into a web app, lenders would be required to collect documents, not digits, from customers. Lenders have well established practices for securely collecting information from customers that they enter in a web app, but the same level of security for collecting documents transmitted via the Internet does not exist. The consequences of adopting a system to borrowers and lenders could be significant, including putting borrowers at increased risk of identity theft and exposing lenders to increased risk of data breaches and fraud.

F. The Requirement To Collect and Verify Documents Should Permit Innovation By Online Lenders

Although storefront lenders may have easy access to documents such as pay stubs, online lenders do not, and as a result have developed innovative technology and underwriting standards that we believe are more predictive than pay stubs. A “one-size-fits-all” rule that is storefront-centric would have significant adverse effects on the nascent online lending industry that has proven to be a key source of innovation in the financial services industry. Curiously, while the U.S. Department of the Treasury recently issued a request for information on expanding access to

16

credit through online marketplace lending,22 the CFPB, which purports to encourage innovation, seems determined to quash innovation through its proposal.

The proposal should reflect the CFPB’s claims of being a 21st century agency and allow for contemporary alternatives to the collection and verification of paper documents. There is constant improvement in the quality and predictive ability of third party data sources, as well as the speed with which such data can be accessed, and the ease with which the data can be integrated into a lender’s existing underwriting processes. Databases of income and expense information,23 or providers of “modeled” income and expense information can accurately predict a given applicant’s income and expenses based on job description and locale.24 And, new account aggregator tools allow access to a consumer’s bank account information, with the consumer’s consent and at the consumer’s direction, in order to show direct deposits (income) and major outflows (obligations).25 A lender could compare this information against an applicant’s stated income and obligations as a “check” against fraud. Not only are these means of verification more efficient and amenable to the online environment, but they are less susceptible to fraud.

Moreover, the lender should be able to verify borrower information based on the risk that the borrower presents. For example, absent some indication that a consumer may be submitting false income or expense information – such as facially inconsistent application information – lenders should not be required to verify income and expense information. In addition, a lender should not be required to verify the customer’s income and expenses repeatedly when the lender is doing business with a trusted customer who has successfully repaid in the past.

VI. Payment Authorization Does Not Expose Consumers to Additional Risk

The proposal would subject to the residual income and document collection, verification and retention requirements any loan where the lender has authority to debit the borrower’s checking account. The proposal treats these ACH authorizations as equivalent to non-purchase money security interests in automobiles, but these are very different legally and functionally. For instance, unlike a secured loan, consumers may revoke ACH authorizations at any time under the CFPB’s Regulation E.

Consumers want and expect the option to authorize automatic recurring payments, since many other financial products (such as mortgages) and non-financial products (such as utilities) offer

22 80 Fed. Reg. 42,866 (July 20, 2015).23 For example, TALX Corp. maintains a database of employment and income history, known as The Work Number, that can be used by verifiers. 24 For example, Experian Information Solutions has developed income estimator models and debt-to-income models, which are validated using verified income information. Trans Union has developed similar models. Salary.com also contains income statistics for specific occupations and specific regions that can be used to compare against an applicant’s stated income.25

17

this service, but lenders will be less likely to offer automatic recurring payments if doing so would sweep the loan within the definition of covered longer-term loans.

A. Consumers Are in Control of Their Own Accounts

Consumers have control of their deposit accounts, and can revoke authorization or stop payment at any time.

The Electronic Fund Transfer Act and Regulation E prohibit lenders from conditioning an extension of credit on a consumer’s agreement to authorize recurring debits.26 That is, lenders must provide an alternate form of payment to automatic recurring debits for installment loans.

NACHA Rules and Regulation E permit consumers to revoke authorizations, stop payments or charge back payments that they believe to be unauthorized.

The CFPB has not presented any data demonstrating that consumers are not able to protect themselves by revoking authorization or stopping payments, or by using an alternative form of payments. If the CFPB is concerned that consumers do not understand their options, the solution is better disclosure, an educational campaign, or, where necessary, supervisory or enforcement actions, not the imposition of heightened underwriting and paperwork requirements.

B. Providers Should Not Be Discouraged From Offering The Convenience Of Recurring Payments To Consumers

Authorization of recurring payments ultimately is a convenience to the consumer and should not be discouraged. Many consumers want and expect the option to authorize automatic recurring payments to lenders. Many other financial services (such as mortgage loans and insurance premiums) and non-financial products (such as utilities, and telecommunications providers and fitness clubs) offer this service.

As noted, lenders will be less likely to offer automatic recurring payments if doing so would sweep their loans within the definition of covered longer-term loans, and subject the loans to overly burdensome underwriting requirements. The CFPB should not restrict consumers’ choice or discourage financial institutions from offering payment options that consumers expect, want, and find convenient.

C. Consumer-Authorized Debits Do Not Receive Preferential Treatment

Contrary to the CFPB’s assertion, authorization to debit a borrower’s account does not provide non-bank lenders a “preferred means of collecting on a loan” ahead of the consumers’ other obligations. Non-bank lenders incur the same risk of non-payment as all other ACH debit originators. Lender ACH debits have no special account access priority over other debits and electronic presentments of checks converted by banks to ACH debits and presented against a consumer’s account that day.

26 See 12 C.F.R. § 1005.10(e).

18

ACH payments are processed in batch, meaning that all ACH debits and credits received before 2:00 AM ET on the prior day are processed at 8:00 AM ET the next morning in the order determined by the receiving depository financial institution. Non-bank lenders cannot be sure when an account debit will be posted or when a paycheck or other credits will be directly deposited to a consumer’s account. Whereas banks and credit unions take payment for their fees as soon as a direct deposit of a paycheck or other credits hit a consumer’s account, non-bank lenders do not have this visibility or ability to time payments. Non-bank lenders stand in line with the consumer’s other creditors and obligations, and can neither verify nor confirm the number and amount of other obligations or promises to pay to which the applicant has committed. The number of other competing payments – including utilities, fitness clubs, prearranged bill payments, checks being presented for payment, and the depository institution itself – and the order of posting those competing payments are beyond the control of the non-bank lender.27

D. Effect of Overdraft Protection

While some depository institutions may apply overdraft protection to ACH debits from consumers’ deposit accounts, payment of overdrafts resulting from ACH debits is discretionary, not automatic and is not subject to consumer opt-in.28 Lenders have no way of knowing whether their borrowers will rely upon discretionary overdraft protection for repayment when they agree to debit the borrowers’ accounts for loan payments. Because small dollar borrowers are typically borrowing $500 to $1000 and may not have other options available, we do not believe that they are likely to be good candidates for discretionary overdraft protection.

Moreover, the available data are inconclusive – although some researchers believe that there is an increased incidence of overdrafts among borrowers of small dollar credit, it would be erroneous to conclude that the small dollar borrowing “caused” any particular overdraft.29 A

27 See National Consumer Law Center, Consumer Banking and Payments Law, at § 2.6.3.9 (discussing bank policies with respect to ordering of transactions, state law considerations, banking agency guidance, and related class action litigation).28 See National Consumer Law Center, Consumer Banking and Payments Law, at § 2.6.3.6 (explaining that overdraft programs are typically structured as “courtesy” or discretionary programs, where the bank avoids an explicit promise to pay an overdraft, so as to not be subject to the finance charge disclosure requirements of TILA and Regulation Z).29 See Pew Charitable Trusts, Overdraft Frequency and Payday Borrowing, at 3 (February 2015) (finding a higher usage of “payday” loans among high-frequency overdrafters, but highlighting that “this analysis cannot conclude whether payday borrowing causes overdrafts or vice versa, but it is clear that the same consumers are using the two products”). We note that the incidental connection between payday loan usage and overdrafts identified in the Pew study may be overstated. The Pew survey only asked consumers about payday loans, but did not ask questions about other authorized debits that might have contributed to a higher frequency of overdrafts. See id. at 6-7 (listing variables considered in low-frequency and high-frequency overdrafter models).

19

consumer’s financial distress may have caused the overdraft, and his or her use of small dollar loans may represent an effort to ameliorate that distress. In fact, we believe it is more likely that small dollar borrowing mitigates or reduces overdraft events by providing the consumer with the means to pay other obligations and maintain cash in his or her deposit account to avoid overdrafts.

In any event, because the order of posting is within the sole discretion of the bank, it is impossible to determine which of the several debits to the consumer’s account caused the consumer to overdraw the account. It also is impossible to know whether the obligations which the consumer satisfied with the proceeds of the small dollar loan would have caused even more overdrafts if not paid in a timely way.

If online lenders are being paid by overdraft protection, surely many other obligations are as well. If certain consumers using overdraft protection in an unhealthy way or if certain banks are being too generous with the extension of discretionary overdraft protection, the answer is to reform overdraft protection, not to risk eliminating the electronic payment option for millions of responsible borrowers.

E. Timing Of Payments Protects Consumers

According to the CFPB, “[a]uthorization to obtain repayment from the consumer’s bank account or wages gives lenders the ability to time and initiate payments to coincide with expected income flows into the consumer’s account.” However, there is nothing nefarious about seeking payment from a consumer when the consumer gets paid. Lenders frequently ask borrowers when they expect to get paid, and allow borrowers to designate payment due dates. Lenders frequently attempt to time payment due dates or dates for automatic debit to coincide with inflows to the borrower’s deposit account. Not only does this help to ensure that the lender gets paid as expected, but it is a convenience to the borrower to have his or her account debited when funds are available, rather than at some other point when the availability of funds may be less certain.

F. New NACHA Rules Will Change The Payments Landscape

The new NACHA rules, effective September 2015,30 should address the CFPB’s consumer protection concerns. The new NACHA rules will provide for a 15% total return rate and a 3% rate of administrative returns to remain a participant in the ACH payments system. Return rates exceeding these thresholds will trigger an eight-step investigation of the originating bank’s ACH activity. We believe that, in practice, originating banks are not likely to initiate transactions for non-compliant originators exceeding these thresholds.

The CFPB should not propose any new requirements on longer-term loans until the new NACHA requirements are fully implemented and their impact is understood. NACHA is comprised of experts in the payments system, and the CFPB staff should not substitute its judgment for that of the NACHA experts before NACHA’s rules are even implemented.

30 NACHA has established a website explaining the rule changes, located at https://www.nacha.org/rules/ach-network-risk-and-enforcement-topics.

20

NACHA has also reduced the limit on returns for unauthorized transactions from 1.0% to 0.5% to remain a participant in the ACH payments system. Unauthorized return rates exceeding this threshold will trigger a risk investigation or enforcement proceeding.

These rules will have a direct and dramatic effect on the behavior of non-bank lenders seeking to collect payments from consumers’ accounts. Lenders will be incentivized to apply even more sophisticated underwriting criteria to ensure they meet these new total return rate thresholds, and to monitor and manage their returns much more effectively, including by taking measures to ensure that consumers have sufficient funds and by limiting re-presentment after failed payment attempts.

Moreover, existing NACHA rules are robust, with respect to the presentment and assessment of fees. Current NACHA rules require that lenders can only re-present a payment twice after an initial failed attempt. In addition, NSF fees must be separately batched from the re-presented payments – i.e., an NSF fee must be presented as a separate debit entry.31 Moreover, any debit entry for an NSF fee must be appropriately authorized under Regulation E.32

G. Debit Cards Payments Are Different Than ACH Payments and Should Not be Subject to the Covered Loan Requirements

The CFPB rule should not apply to payment authorizations from a consumer’s debit card. Unlike an ACH debit, if funds are not available from a consumer’s debit card, the transaction will not be authorized (unless the consumer expressly opts in to overdraft protection), and will not be returned NSF. Consumers do not pay a fee when a debit card transaction is denied.33 The CFPB should not regulate as “covered longer-term loans” those loans where the consumer has provided a debit card authorization, or otherwise is not penalized by his or her bank for failed payment attempts.

VII. Loan Limits Do Not Protect Consumers

Limiting the ability of lenders to lend to repeat customers does not help consumers – it just increases loan acquisition and origination costs and makes it more difficult for lenders to predict ability to repay. Moreover, if the CFPB’s proposed ability to repay requirement performs as intended, then limits on re-borrowing or prohibitions on having multiple loans outstanding would be unnecessary and superfluous.

31 NACHA Operating Rules § 2.14.3 (requiring that Originators submit Return Fee Entries “as a separate batch that contains the words ‘RETURN FEE’ in the Company Entry Description field of the Company/Batch Header Record”). 32 12 C.F.R. § 1005.10(d) (provisions regarding required notice of “transfers varying in amount”).33 In addition, card networks, similar to NACHA, regulate the unauthorized return rates of participants on their networks.

21

Customers who have paid in the past are the ones who are most likely to pay in the future, and all financial services – insurance, credit cards, auto loans, securities – depend on repeat business for profitability, because of the high costs of acquisition, underwriting, fraud and credit losses. The ability to loan to existing customers improves underwriting outcomes, and minimizes acquisition and underwriting costs, as well as fraud and credit loss.

The CFPB’s proposal will also make it much more difficult for borrowers to improve their credit, and thereby get access to lower cost credit. The best way for consumers to establish a positive credit history is to allow them to make consistent, successful loan repayments.

In addition, the increased acquisition costs that would result from the imposition of loan limits are more easily borne by big companies, thereby disadvantaging smaller lenders and startups.

Furthermore, imposing loan limits penalizes consumers for repaying their debts, which creates perverse incentives for a consumer protection rule. If consumers are only permitted to borrow twice per year, they will quickly adapt to those limits and end up borrowing more money than they need for a longer period of time, and carrying more debt and for a longer period than they would in the absence of loan limits. In addition, consumers may not pay back their loans as quickly as possible, because they will not be given credit for prepayment and will continue to be barred from re-borrowing.

The exception to the loan limits in the proposal for a change in a consumer’s financial circumstances will have little utility in the real world. If the consumer’s financial situation improves, it is less likely he or she would need a second or third loan at all.

The proposal will frustrate consumers with its complexity and inflexibility. Consumers may find that government-mandated waiting periods for small loans are an unnecessary and intrusive burden, especially when other aspects of their financial lives, including credit cards, student loans and overdrafts, do not include such restrictions. Consumers have developed relationships with trusted lenders – why should they no longer be able to borrow from those lenders and why should they be required to search for inferior alternatives?

A. Flexibility in Loan Limits

We support the CFPB’s “off-ramp” proposals in principle, and believe a simplified off-ramp would be more effective than blunt loan limits. If the CFPB is concerned about borrowers continually rolling over their loans, instead of a flat restriction on reborrowing in a sequence, the rule could require the consumer to pay down a specified percentage (e.g., 25%) of principal before reborrowing.

However, assuming that the CFPB’s eventual proposal includes some form of loan limits, it is important that these limits be flexible. If a consumer obtains a loan and pays it back in full without a refinancing, rollover, or taking out another loan, the consumer should be free to obtain another loan. Consumers should not be entirely locked out of the market for short-term credit because sometimes unexpected events arise that create the need for emergency borrowing. Customers cannot predict when they will need to borrow money short-term. There is no reason

22

why a consumer who receives an unexpected $200 medical bill should be precluded from borrowing to meet that expense simply because she successfully borrowed and repaid another loan 45 days earlier.

The proposal also might allow for a borrower’s demonstrated need – where the borrower or lender can demonstrate that the use of the loan proceeds of the loan outweighs the cost of the loan. That is, if a borrower has had three loans and is in the proposed 60-day waiting period, the borrower should be able to present his or her case that the need for credit is justified because the benefit of the proceeds of the loan exceeds the cost of the loan. For example, if the consumer must pay for a car repair that is critical for the consumer’s ability to get to work and earn a living, then forbidding that consumer to access to credit would create much greater harm than the cost of the additional loan.

VIII. The Federal Database Proposal Is Not Workable

The CFPB proposal relies on the development of private databases, which will not occur if the CFPBs own impact assumptions are accurate. For example, the proposal relies upon the market responding with specialized credit bureaus that meet CFPB standards to fulfill the database requirement – i.e., to serve as a repository for lending information, so that lenders can ensure that cooling off and similar requirements are observed – but the CFPB estimates that the proposal will eliminate three quarters of the market, meaning there would be few surviving lenders to whom a service provider could sell such a product. In addition, the proposal would require lenders to report loan history to all specialized credit bureaus, but would only require them to obtain a credit report (i.e., for a fee) from a single one of these credit bureaus. If the CFPB is correct about the economic effect of its proposal, it would not be economic for any private enterprise to develop a specialized credit bureau to provide the required database services, where the credit bureau will be required to accept data from all comers but only actually sell credit reports to a fraction of those companies.

Moreover, if no private-sector database develops or the CFPB does not approve any database, then no loans can be made because no lender can comply with the CFPB rule. We believe that that would be a strange contingency on which to obliterate an entire credit market. Even if a database is established, it will take a considerable amount of time to build a consumer reporting agency that would be compliant with the CFPB standards, obtain CFPB approval to begin operations, develop the systems needed to collect the loan performance data from lenders, integrate the systems into lenders loan management systems, determine how to report loan performance data to lenders, and collect sufficient data to be useful.

In addition, database requirements disadvantage smaller lenders. The requirement to furnish data to all CFPB-certified credit bureaus – whether a single credit bureau or 50 credit bureaus – imposes a disproportionate burden on small business. There is an enormous fixed cost associated with developing the procedures, systems, training and technology associated with communicating detailed consumer data to different credit bureaus, each of which provides a different API and method of furnishing. Each lender will have to make its systems communicate with the systems of each of the credit bureaus that are approved by CFPB for the receipt of

23

lender data. Larger lenders will be able to spread these development costs over a much larger number of loans, for a lower per-unit compliance cost.

Finally, mandatory furnishing of data to credit bureaus is antithetical to the voluntary nature of the credit reporting system. The Fair Credit Reporting Act does not mandate the furnishing of information to credit bureaus, and instead is predicated on a system in which banks and other creditors voluntarily furnish information to credit bureaus.34 Creditors may elect to furnish or not furnish information to credit bureaus based on what they believe to be in the interests of their customers or their own interest. The voluntary nature of the credit reporting system is one of the primary reasons for the dynamism of U.S. credit markets.35

IX. The Evidence Shows that the Market for Short-Term Loans Works

We believe that the evidence shows that consumers are not injured by short-term loans, but are in fact helped by such loans and are able to understand the risks of such loans and make informed choices.

Sustained use of credit is not itself an injury. Credit bureau data shows that borrowers who engage in protracted refinancing actually have better financial outcomes (measured by changes in credit scores) than consumers whose borrowing is limited to shorter periods.36

Even consumers who default on payday loans do not necessarily sustain an injury. Data shows that the credit score changes of borrowers who default on payday loans differ immaterially from the score changes of borrowers who do not default on payday loans.37

Moreover, there is no reason to think that borrowers of covered loans generally suffer from infirmity, illiteracy or ignorance. In fact, recent research shows that borrowers of covered loans are increasingly well-educated and sophisticated with respect to their use of financial services, and that there is no reason to believe that these borrowers are vulnerable or otherwise unable to protect themselves:

34 See CFPB, Key Dimensions and Processes in the U.S. Credit Reporting System: A review of how the nation’s largest credit bureaus manage consumer data, at 15 (December 2012); Board of Governors of the Federal Reserve System, Report to the Congress on Credit Scoring and its Effects on the Availability and Affordability of Credit, at 13 n.12 (August 2007); Robert B. Avery, Raphael W. Bostic, Paul S. Calem, and Glenn B. Canner, An Overview of Consumer Data and Credit Reporting, 89 Fed. Res. Bull. 47-73 (Feb. 2003).35 See Michael G. Oxley, Opening Statement, HR 2622, The Fair and Accurate Credit Transactions Act (July 9, 2003) (“[T]he ‘miracle of instant credit’ created by our national credit reporting system has given American consumers a level of access to financial services and products that is unrivalled anywhere in the world”).36 See Jennifer Priestley, Payday Loan Rollovers and Consumer Welfare¸(Dec. 5, 2014) .37 See Ronald Mann, Do Defaults on Payday Loans Matter? (Dec. 1, 2014).

24

A recent study by the Urban Institute, a non-profit, non-partisan think tank, found that households that use alternative financial services are increasingly from middle- and upper- income ranges.38 These borrowers are more likely to have completed some higher education, and are better able to understand the terms and consequences of credit products.

In addition, the median age of borrowers of small dollar single pay loans and installment loans is 41 and 43, respectively.39 Most borrowers are not young or elderly.

A study by Ronald J. Mann of Columbia University found that most borrowers of payday loans not only expect that they will continue borrowing for some time after the initial loan, but also can predict within one period when they would be free of debt.40 These results “undermines the notion (characteristic of much of the legal and policy literature on the subject) that the repeated borrowing that is typical of payday borrowers generally reflects surprise on the part of the borrowers or deception on the part of the lenders.”

The CFPB’s own study found that nearly half of new borrowers have only one loan sequence during the year, and a majority of short term borrowers use two or fewer loans in a sequence.41

X. The 36% Military APR Trigger For Rule Coverage Is Not in the Public Interest

If the CFPB is concerned about authorization of debits somehow causing consumers to take out “unaffordable” loans, why are the residual income and document verification not extended to all loans, including those with less than 36% APR? Affordability is not just a function of the APR, but also depends upon the principal amount, loan term, and, ultimately, periodic payment amount.

The CFPB has not articulated or justified a consumer protection rationale for singling out unsecured installment credit with greater than 36% APR for these onerous restrictions.

The uneven regulation of credit products may cause installment lenders to exit the market or take other steps to circumvent these burdensome rules and leave consumers with inferior credit alternatives.

A. APR Is a Flawed Measurement, and Military APR Is Doubly Flawed.

An APR expresses the cost of a loan as an annualized rate, which makes no sense for a short-term loan. The use of APR in this context could cause consumers to choose more expensive alternatives that do not suit their needs simply because they appear to be lower cost, as the result

38 See Gregory B. Mills, As the Recovery Progresses, Use of Nonbank Credit Rises (Mar. 9, 2015). 39 Bretton-Woods, Inc., The State of Online Short-Term Lending: Statistical Analysis Report, at 3 (July 2015).40 See Ronald J. Mann, Assessing the Optimism of Payday Loan Borrowers, 21 Sup. Ct. Econ. Rev., 105, 118 (2013). 41 Consumer Financial Protection Bureau, Data Point: Payday Lending (Mar. 2014) .

25

of a lower APR. The Federal Reserve Board has found that APRs confuse consumers, and either eliminated effective APR disclosures or diminished the prominence of APR disclosures for purposes of certain Truth-in-Lending Act private student loan disclosures under Regulation Z.42

The CFPB has compounded this error by proposing the use of an “all-in” or “Military” APR to calculate the interest and fees subject to the 36% rate cap. The Department of Defense adopted the Military APR to regulate credit to servicemembers under the Military Lending Act. The Military APR calculation requires consideration of several types of fees not traditionally covered by state usury limits: application fees, finance charges, credit insurance premiums, and fees for non-credit products that are “ancillary” to the credit.43

Military APRs include fees that are unrelated to credit and result in an arbitrary and distorted measurement of the cost of credit. The use of the all-in APR can cause even a small, routine fee to result in a highly inflated figure that far exceeds an annualized 36%. As evidence, even payday loan alternatives encouraged and endorsed by the CFPB and consumer advocacy groups could exceed the CFPB’s rate cap because of their fees:

Payday Alternative Loans. The National Credit Union Administration encourages credit unions to offer a “Payday Alternative Loan,” a low cost, small-dollar loan. Loans range from $200 to $1,000, with application fees not to exceed $20 and an interest rate not to exceed 28% APR. Terms range from one to six months. However, when the application fee is included in the APR calculation, many of these loans would exceed the CFPB’s 36% rate cap. For example, a one month $500 loan with a $20 application fee and 28% annual interest would reach an all-in APR of more than 70% and a $200 loan a nearly 150% all-in APR. Therefore, the NCUA itself has expressed concern that the rate and fee for many Payday Alternative Loans would be higher than a 36% all-in APR cap.44

FDIC Small Dollar Lending Program. Under the FDIC’s 2007 Affordable Small-Dollar Loan Guidelines, banks were encouraged to offer affordable small-dollar loans. The guidelines promoted loans under $2,000 with APRs no greater than 36%, plus reasonable fees. Therefore, loans under the FDIC’s guidelines could violate the CFPB’s rate cap.

42 See 74 Fed. Reg. 43,428, 43,508 (Aug. 26, 2009) (“The quantitative consumer research conducted by the Board validated the results of the qualitative testing; it shows that most consumers do not understand the effective APR, and that for some consumers the effective APR is confusing and detracts from the effectiveness of other disclosures.”); 74 Fed. Reg. 41,194, 41,196 (Aug. 14, 2009) (“Consumer testing of private education loan disclosures . . . confirmed that most consumers understand the interest rate and that it is one of the most important terms to them. At the same time, most consumers do not understand the APR and incorrectly believe that the APR is the interest rate.”).43 12 C.F.R. § 232.3(h)(1); see also 80 Fed. Reg. 43,560, 43,608 (July 22, 2015).44 National Credit Union Administration, Proposed Defense Department Lending Rule Could Affect Credit Unions, Members (Sept. 26, 2014).

26

Furthermore, when interest accrues for a shorter period, APRs are higher, even though the borrower pays exactly the same amount, and the lender’s risk justifies the same profit margin as a longer term loan. In other words, the duration of a loan is a key driver of the APR calculation and makes shorter-term loans appear more costly than longer-term loans, all else being equal.

The following chart shows how an identical low dollar fee results in a dramatically higher all-in APR as the duration of a loan decreases:

These examples show that the economics of short-term, small dollar lending are not well-represented by APRs. Thus, according to one commentator, imposing an APR cap on a shorter term loan “makes as much sense as taking the $22 cab fare from the Los Angeles Airport to Hermosa Beach, CA (a 7 mile trip), and calling it exploitation because at that rate it would cost over $6,500 for the cab ride from Los Angeles to Montgomery, AL, when a flight runs in the $600 range.”45

B. Small Dollar Loans Cannot Be Made Economically at 36% APR

The CFPB’s proposal ignores the economic viability of the types of loans it is proposing to regulate with its burdensome new ability to repay/residual income and loan limit regulations. Loans under the Payday Alternative Loan program and FDIC Small Dollar Loan Program have proven to be unprofitable, even though credit unions and banks have access to low cost NCUA-insured and FDIC-insured deposit funding, as well as direct access to share and deposit accounts, which can be debited for repayment. (Moreover, credit unions enjoy a valuable tax advantage.46)

45 Prof. Daniel J. Smith, Restrictions on Payday Lending are Unfair and Harmful to Borrowers (Feb. 6, 2015).46 Kenneth J. Kies and Bert Ely, Tax Exemption for Credit Unions: An Unjustifiable $10 Billion Tax Expenditure (discussing credit unions’ exemption from federal income tax).

27

The National Association of Federal Credit Unions has stated that Payday Alternative Loans are “loss-leaders in credit unions and are offered strictly for the benefit of their members who are in need of short-term small-dollar alternatives to payday lenders.”47

The FDIC has conceded that “the interest and fees generated [from FDIC Small Dollar Loans] are not always sufficient to achieve robust short-term profitability. Rather, most pilot bankers sought to generate long-term profitability through volume and by using small-dollar loans to cross-sell additional products.”48

Moreover, the CFPB’s 5% GMI loans also cannot be made profitably. This option would prevent a consumer who has an annual gross income of $30,000 – the mean income reported by credit bureaus for borrowers of single payment loans49 – from taking out a 30 day loan that exceeds $125. At that scale, lenders will not be able to generate revenues to remain in business.

There is both a lack of research and a lack of real-world evidence that small-dollar unsecured loans are economically viable at rates of 36% or less, even for tax-exempt, non-profit credit unions. If these loans are not economically viable at lower rates, then high-risk consumers living paycheck to paycheck may be denied access to credit necessary to help them to meet unexpected emergencies. It is not clear how and by whom the CFPB expects this demand to be met.

Small dollar lenders have legitimate reasons for charging more than a 36% all-in APR. For example, small dollar lenders need to charge relatively higher rates of interest and fees because they provide credit to subprime borrowers ignored by banks, and who are more likely to default than the average borrower. As a result, even charging higher rates of interest, payday lenders have very low profit margins. Pure payday lenders (excluding pawn shops) have average profit margins of only 3.57%.50 Therefore, “contrary to conventional wisdom, these firms fall far short of profits for mainstream commercial lenders.”51

With low profit margins, non-bank lenders cannot survive a 36% rate cap. As an example, a state-licensed payday lender frequently charges $15 per $100 borrowed, due on the borrower’s next payday. At a 36% APR, the payday lender would be prohibited from charging more than

47 National Association of Federal Credit Unions, Comment Letter to Department of Defense Proposed Rule on Limitations on Terms of Consumer Credit Extended to Service Members and Dependents (Dec. 23, 2014) (emphasis added).48A Template for Success: The FDIC’s Small-Dollar Loan Pilot Program , FDIC Quarterly 4 No. 2 (2010).49 Bretton-Woods, Inc., The State of Online Short-Term Lending: Statistical Analysis Report, at 3 (July 2015).50 Aaron Huckstep, Payday Lending: Do Outrageous Prices Necessarily Mean Outrageous Profits?, 12 Fordham J. Corp. & Fin. L. 203, 227 (2007).51 Id.

28

$1.50 per $100 borrowed. With already low profit margins, suffering a 90% loss in revenue is clearly not economic.

Moreover, fixed costs for small-dollar loans can be substantially similar to those for higher amount loans, but the interest earned is much less because of the lower loan amount of the loan. For example, 10% interest per year on a $3,000 loan will earn $300 and on a $500 loan $50. If the lender’s fixed costs are over $50, the $500 loan is unprofitable, so it likely will not be offered or made.52

C. The CFPB’s Rate Cap Will Hurt Consumers

As Federal Trade Commission commissioner and economist Joshua D. Wright has recognized, “Economic research has overwhelmingly concluded that usury ceilings generally harm those that they are intended to help.”53

That is because usury limits restrict access to legal credit, but do not eliminate demand for credit. In the absence of available and convenient small dollar credit, consumers will instead use higher-risk, more expensive and inferior alternatives, such as late payment fees, utility service reconnection fees, bounced checks and overdraft fees.

In fact, the proposal’s rate cap and other requirements may also encourage lenders to charge higher interest rates to consumers who have already been deemed to have an ability to repay their loans, in order to recoup the revenues foregone and costs incurred due to the proposal’s requirements.

XI. The Proposal Would Reduce Competition, Innovation, Consumer Choice, and Access to Credit

The proposal would create a barrier to entry in the consumer lending industry, because of the higher fixed costs of building a compliance system.54 The proposal requires lenders to invest in systems to access, collect, process, store, verify and report consumer loan data. The proposal also requires paperwork to document “ability to repay” determinations. As a result of these high fixed costs, the proposal favors larger players, which have a lower per unit cost of compliance.

52 See Department of the Treasury, Public Input on Expanding Access to Credit Through Online Marketplace Lending, 80 Fed. Reg. 42,866, 42,867 (July 20, 2015) (stating that “it can cost the same amount to underwrite a $300 consumer loan as a $3000 loan”). 53 Comment of Commissioner Joshua D. Wright on the Department of Defense’s Proposed Amendments to its Regulation Implementing the Military Lending Act (Dec. 26, 2014).54 See The Cost of Bank Regulation: A Review of the Evidence, by Gregory Elliehausen, Staff, Board of Governors of the Federal Reserve System, April 1998 at 26-27 (stating that “generally, the larger the institution, the lower the costs” of developing and maintaining a compliance program). This same study concluded that in 1991, “regulatory costs account[ed] for 12 percent to 13 percent of [banks’] noninterest expenses,” or $15.7 billion in regulatory expense as compared to $125.9 billion of non-interest expenses overall. Id. at 29.

29

Companies that have the resources to develop systems to collect and verify borrower information will have a significant competitive advantage over smaller lenders.

A. The Proposal Could Cause A Migration To Inferior Alternatives

The most recent FDIC National Survey of Unbanked and Underbanked Households reported that 25% of households had bank accounts but also used alternative financial services, including small dollar loans.55 As a result of reduced access to credit, some of these borrowers may migrate to alternatives that are inferior to covered loans to meet their needs. Alternatives can be more expensive and inconvenient, including defaults on other bills, incurring late fees that are usually higher than fees and interest rates on covered loans; bounced checks; overdrawn bank accounts; and the cut-off and reconnection of utility services, which can be very expensive, inconvenient, and can be dangerous (i.e., living in a house with no heating, cooling, water, electricity, or telephone line).56 Some of these alternatives can also damage the consumer’s credit rating, or jeopardize the consumer’s deposit account.

A study by the Federal Reserve Bank of New York found that “Georgians and North Carolinians do not seem better off since their states outlawed payday credit: they have bounced more checks, complained more about lenders and debt collectors, and have filed for Chapter 7 (“no asset”) bankruptcy at a higher rate” and that “[f]orcing households to replace costly credit with even costlier credit is bound to make them worse off.”57

Similarly, a study of survey data by Jonathan Zinman of Dartmouth College found that new binding restrictions on loan terms in Oregon caused a migration to “plausibly inferior substitutes,” and caused Oregon respondents to be significantly more likely to experience an adverse change in financial condition.58

Many borrowers of covered loans may not have access to other loans, and the Bureau does not explain where these borrowers will be able to turn for help.

B. Innovators Are Often Higher Cost Providers

Innovators often cannot compete on cost, particularly when larger providers are able to spread compliance costs over a larger number of units sold. Innovators are able to compete on other

55 FDIC, 2013 National Survey of Unbanked and Underbanked Households, at 8-9.56 See Donald P. Morgan and Michael R. Strain, Payday Holiday: How Households Fare After Payday Credit Bans , at 1 (Nov. 2007) (“[P]ayday credit is cheaper than overdraft credit for overdrafts below $180”); Bretton-Woods, Inc., Online Short-Term Lending: Statistical Analysis Report, at 8 (Feb. 28, 2014) (demonstrating that average overdraft fees are be more expensive than small dollar loans in filling a consumer’s credit need).57 Donald P. Morgan and Michael R. Strain, Payday Holiday: How Households Fare After Payday Credit Bans (Nov. 2007).58 Jonathan Zinman, Restricting Consumer Credit Access: Household Survey Evidence on Effects around the Oregon Rate Cap (Oct. 2008).

30

elements of the transaction – such as speed, safety, and convenience – that may matter more than price to some consumers.59 This proposal is based entirely on price, and fails to recognize other aspects of the transaction. In so doing, the proposal disadvantages innovators and stifles innovation.

XII. Alternative Solutions

As noted by Pew and others, the CFPB’s proposal is needlessly complex and could benefit from considerable simplification.

In addition, as outlined in more detail above, the proposal should:

Allow for alternatives to the residual income and household budgeting requirement, such as the use of automated algorithms to predict likelihood of repayment;

Provide for CFPB review of underwriting models to determine, after the fact, how well those models predict the risk of nonpayment, rather than prescribing a vague and unproven residual income test that will needlessly disqualify creditworthy borrowers;

Allow for alternatives to collection and verification of paper documents, including the use of commercially available databases and “modeled” data, as well as online account access tools;

Allow for lenders to take a “risk-based” approach to document verification – rather than requiring verification for all applicants, even those who present a very low risk of fabricating income or expense information;

Not impose limits on the number of loans per year or the number of loans outstanding at any one time, particularly for longer-term loans;

Allow for exceptions to those loan limit requirements beyond the proposed showing of “changed circumstances,” including exceptions for demonstrated need and exceptions where consumers are regularly paying down principal with each payment.

Allow for NACHA Rules to be fully implemented and determine their effect on the market before making new rules regarding loans repaid by direct debit;

Not treat loans differently based on payment method or APR; and

Not apply to debit card payments, which present no risk of NSF fees, as compared to ACH debits.

Furthermore, the CFPB needs to clearly articulate the harm it is seeking to prevent and demonstrate how its proposals will remedy that harm. More specifically, if the CFPB is

59 American Banker, Five Reasons You Can’t Ignore the Neobanks (Apr. 27, 2015).

31

concerned about consumer authorization of recurring payments, the CFPB should address payment authorization directly, rather than by imposing burdensome and arcane residual income and document verification requirements on lenders. OLA would like to work with the CFPB to develop targeted, common sense regulation of payment authorizations instead of the proposal’s overbroad, indirect regulation that will unnecessarily shrink the market for unsecured installment loans. For example:

Lenders should offer consumers the right to rescind a loan by the end of the next business day following approval, with no fees or other obligations.

Lenders should permit consumers to change the payment amount (i.e., to pay down additional principal) within 24 hours after an account debit occurs.

We support a requirement to provide the borrower with payment reminders, but lenders should be allowed to provide notice through simple and concise emails or texts (rather than the lengthy list of information the CFPB is proposing to require). Payment reminders should be provided 24 hours prior to the ACH revocation deadline.

The CFPB should conduct an educational campaign alerting consumers to their rights to revoke authorization, stop payment, recover for fraudulent payments, and otherwise control what happens with respect to their deposit account.

Finally, if the CFPB is concerned that longer-term loans are somehow causing consumers to overdraw their accounts, we would submit that, given the multiple and variable nature of payments, determining which payment “caused” an overdraft is impossible, and, in any event, is beside the point – the cause of an overdraft is not any particular debit, but too many debits against an insufficient balance. If the CFPB’s concern is excessive overdrafts, the proper way to address that is through reform of the overdraft rules, not by restricting the ability of consumers to authorize recurring payments.

* * *

32

We appreciate the opportunity to provide input on this important regulatory initiative. Thank you also for the time that you and your staff have devoted to building an understanding of this market and the consumers that we serve. If you have questions or need additional information, please feel free to contact me at lmcgreevy@oladc.org.

Very Truly Yours,

Lisa S. McGreevyPresident and CEO

33