The July 2017 Senior Loan Officer Opinion Survey on Bank Lending Practices
Table 1 | Table 2 | Chart data
Table 1 (PDF) | Table 2 (PDF) | Charts (PDF)

The July 2017 Senior Loan Officer Opinion Survey on Bank Lending Practices (SLOOS) addressed changes in the standards and terms on, and demand for, bank loans to businesses and households over the past three months.¹ This summary discusses the responses from 76 domestic banks and 22 U.S. branches and agencies of foreign banks.²

Regarding loans to businesses, the July survey results indicated that, on balance, demand for commercial and industrial (C&I) loans weakened over the second quarter of 2017 while banks left their standards on C&I loans basically unchanged.³ Survey respondents also reported that standards on commercial real estate (CRE) loans tightened while demand weakened on net.

For loans to households, banks reported that lending standards on all categories of residential real estate (RRE) loans eased or remained unchanged, although the net share of banks reporting easing was, at most, moderate in each category. Banks also reported, on net, that demand for most categories of RRE loans strengthened over the second quarter, although, again, the net share was never more than moderate. In addition, modest net fractions of banks reported tightening standards and weaker demand for auto and credit card loans.

Responses to a set of special annual questions on the approximate levels of lending standards suggested that domestic banks' lending standards for all categories of C&I loans are currently easier than the midpoints of the ranges that have prevailed since 2005 (explained more fully later). In contrast, banks also indicated that standards on all types of CRE loans are currently tighter than the midpoints of their respective ranges.

In addition, on balance, banks reported that the levels of standards for all types of RRE loans are currently at least as tight as the midpoints of the ranges observed since 2005. Moreover, banks indicated that the levels of standards for consumer loans to subprime borrowers are currently still tighter than the midpoints of their respective ranges, while the analogous readings for consumer loans to prime borrowers are currently easier than their midpoints.

Lending to Businesses
(Table 1, questions 1-12; Table 2, questions 1-8)

Questions on commercial and industrial lending. On balance, banks reported that standards for C&I loans were basically unchanged over the past three months for loans to both small firms and large and middle-market firms.⁴ However, terms on C&I loans became less restrictive, on balance, with specific loan terms all either easing or remaining basically unchanged. Specifically, a significant net percentage of banks reportedly narrowed spreads of loan rates over the cost of funds, while a moderate net share of banks reportedly increased the maximum size of credit lines and decreased the use of interest rate floors for large and middle-market firms. A modest net percentage of banks reported easing these terms to small firms as well. Besides a few other terms for large and middle-market firms that were modestly eased, other terms remained basically unchanged on net.⁵

Among the domestic respondents that reportedly eased standards or terms on C&I loans over the past three months, more aggressive competition from other bank or nonbank lenders was by far the most emphasized reason for easing. In particular, a majority of banks reported that more aggressive competition was an important reason for easing, with almost five times as many banks identifying the reason as "very important" as any other reason.

Regarding the demand for C&I loans, a moderate net share of domestic banks reported that demand from large and middle-market firms weakened, while a modest net share of banks reported that demand from small firms did so. The reported reasons for weakening loan demand were less concentrated than the reasons for having eased standards. Each of the following reasons for weaker demand was cited by at least half of the banks that reported weaker demand: shifts in customer borrowing to other bank or nonbank sources and decreases in customers' needs to finance inventory, accounts receivable, investment in plant or equipment, and mergers or acquisitions.

Meanwhile, foreign banks reported that C&I lending standards and demand remained basically unchanged, on balance, in the second quarter of 2017. Changes in C&I loan terms were mixed; a moderate net share of banks reported having narrowed loan spreads and reduced the use of interest rate floors, but a modest net share reported decreasing the maximum size of credit lines.

Questions on commercial real estate lending. On net, domestic survey respondents indicated that their lending standards for all major categories of CRE loans tightened during the second quarter. In particular, a moderate net fraction of banks reported tightening standards for construction and land development loans and loans secured by multifamily residential properties, while a modest net share of banks reported tighter standards for loans secured by nonfarm nonresidential properties.

Banks also reported that demand for CRE loans weakened during the second quarter. A modest net fraction of banks reported weaker demand for construction and land development loans and loans secured by multifamily residential properties, while demand for nonfarm nonresidential loans remained basically unchanged on net.

Meanwhile, a modest net share of foreign banks reported tightening standards for CRE loans. Also, in contrast to the domestic respondents, a significant net share of foreign banks indicated that demand for CRE loans strengthened in the second quarter of 2017.

Lending to Households
(Table 1, questions 13-26)

Questions on residential real estate lending. On balance, banks reported that standards for all surveyed categories of RRE lending either eased or were unchanged over the past three months.⁶ A moderate net share of banks reported easing underwriting standards for jumbo residential mortgages that do not conform to qualified mortgage (QM) rules, while a modest net share of banks reported easing standards for QM jumbo mortgages and mortgages that are eligible to be securitized by government sponsored enterprises (GSE eligible). Standards for other categories of home-purchase mortgages as well as for revolving home equity lines of credit were basically unchanged on net.

Banks also reported stronger demand for most categories of RRE loans on net. A moderate net share of banks reported stronger demand for QM jumbo mortgages, while a modest net share reported stronger demand for GSE-eligible, government, and non-QM jumbo mortgages. However, a modest net fraction of banks reported weaker demand for non-QM non-jumbo residential mortgages. Demand for other mortgage categories and for home equity lines of credit was basically unchanged on net.

Questions on consumer lending. A modest net share of banks reported tightening lending standards on credit card and auto loans, whereas standards on other consumer loans remained basically unchanged. Regarding terms on consumer loans, modest net fractions of banks reportedly widened spreads of loan rates over their cost of funds in all three consumer loan categories. Additionally, a modest net share of banks reported increasing their minimum required credit score for credit card loans, although a similar net share of banks also reported increasing the limits on their credit cards. Other terms on consumer loans remained basically unchanged.

Banks reported that demand for consumer loans weakened in the second quarter: A modest net share reported that demand for auto and credit card loans weakened, while demand for other consumer loans reportedly remained basically unchanged on net.

Special Questions on the Current Levels of Lending Standards
(Table 1, question 27; Table 2, question 9)

The July survey included a set of special questions that asked respondents to describe the current levels of lending standards at their bank.⁷ Specifically, for each loan category surveyed, respondents were asked to consider the range over which their bank's standards have varied between 2005 and the present and then to report where the current level of standards for such loans currently resides relative to the midpoint of that range.

Domestic banks reported that their current lending standards on all categories of C&I loans remained at levels that are easier than the midpoints of their respective ranges since 2005. A significant net share of domestic banks reported that standards are currently easier than the respective midpoints for non-syndicated loans to both small and large and middle-market firms as well as for syndicated loans to investment-grade firms.⁸ Additionally, a moderate net share of domestic banks reported a relatively easy level of standards for loans to very small firms, while a modest share reported a relatively easy level of standards for syndicated loans to below-investment-grade firms. Moreover, relative to the responses in the July 2016 survey, the level of lending standards appears to have eased, on net, for all categories over the past year.

In contrast, foreign banks reported that the current levels of their C&I lending standards are, if anything, generally tighter than the midpoints of their respective ranges. A significant net share of foreign banks reported that the level of standards is tighter than the midpoint of its range for non-syndicated loans to large and middle-market firms, while a moderate net share of banks reported a tighter level for non-syndicated loans to small firms.⁹ The level of standards for syndicated loans is reportedly around the midpoint of its range on net.

Regarding the levels of standards on CRE loans, domestic banks reported that the current levels of their standards on all major categories of these loans are tighter than the midpoints of the ranges that have prevailed since 2005. A significant percentage of domestic banks reported, on balance, that current levels of standards are tighter than the respective midpoints on loans secured by multifamily residential properties and on loans for construction and land development purposes, while a moderate net percentage reported that levels of standards are tighter than the midpoint on loans secured by nonfarm nonresidential properties. A major net share of foreign banks reported a relatively tight level of standards for construction loans, while a significant net share did so for multifamily and nonfarm nonresidential loans. However, only about half of the foreign banks responded to each question, as foreign banks are a relatively small part of the CRE loan market.

With respect to RRE loans, on balance, domestic banks reported that lending standards for most of the five categories included in this survey remained somewhat tighter than the midpoints of the ranges of those standards since 2005. Subprime residential mortgages remained the category that was most consistently reported as tight, on net, with a significant net share of banks reporting that standards are currently tighter than their respective midpoint. Additionally, a moderate net share of banks reported relatively tight standards on jumbo loans and on home equity lines of credit, while the current level of standards was reported to be around the midpoint, on net, for GSE-eligible and government residential mortgages.

On balance, banks' current levels of standards on consumer loans were reported to be on the tight end of the range since 2005 for subprime borrowers while being somewhat easier for prime borrowers. In particular, significant net fractions of banks reported that the levels of their standards are currently tighter than the midpoints of their respective ranges for both auto and credit card loans to subprime borrowers. However, a moderate net percentage of banks reported that the current level of standards is easier than the midpoint, on net, for auto loans to prime borrowers, while standards are around the midpoint for credit card loans to prime borrowers. A modest net share of banks also reported that the current level of standards is tighter than the midpoint for consumer loans other than credit card and auto loans. However, even for loans to prime customers, banks indicated tightening relative to last year—in all five consumer loan categories, banks reported that the current levels of standards are tighter, on net, than in the July 2016 survey.

This document was prepared by David Glancy, with the assistance of Akber Khan, Division of Monetary Affairs, Board of Governors of the Federal Reserve System.

---

1. Respondent banks received the survey on or after June 19, 2017, and responses were due by July 3, 2017. Return to text

2. Unless otherwise indicated, this document refers to reports from domestic respondents. Return to text

3. For questions that ask about lending standards or terms, "net fraction" (or "net percentage") refers to the fraction of banks that reported having tightened ("tightened considerably" or "tightened somewhat") minus the fraction of banks that reported having eased ("eased considerably" or "eased somewhat"). For questions that ask about loan demand, this term refers to the fraction of banks that reported stronger demand ("substantially stronger" or "moderately stronger") minus the fraction of banks that reported weaker demand ("substantially weaker" or "moderately weaker"). For this summary, when standards, terms, or demand are said to have "remained basically unchanged," the net percentage of respondent banks that reported either tightening or easing of standards or terms, or stronger or weaker demand, is between 0 and 5 percent; "modest" refers to net percentages between 5 and 10 percent; "moderate" refers to net percentages between 10 and 20 percent; "significant" refers to net percentages between 20 and 50 percent; and "major" refers to net percentages over 50 percent. Return to text

4. The survey asked respondents separately about their standards for, and demand from, large and middle-market firms, which are generally defined as firms with annual sales of $50 million or more, and small firms, which are those with annual sales of less than $50 million. Return to text

5. In addition to the already mentioned loan terms, banks were asked about changes in risk premiums, cost of credit lines, maximum maturity, loan covenants, and collateralization requirements. The maximum maturities and loan covenants reportedly eased modestly for large and middle-market firms, while other terms remained basically unchanged. Return to text

6. The seven categories of residential home-purchase loans that banks are asked to consider are GSE-eligible, government, QM non-jumbo non-GSE-eligible, QM jumbo, non-QM jumbo, non-QM non-jumbo, and subprime. See the survey results tables that follow this summary for a description of each of these loan categories. The definition of a QM was introduced in the 2013 Mortgage Rules under the Truth in Lending Act (12 CFR Part 1026.32, Regulation Z). The standard for a QM excludes mortgages with loan characteristics such as negative amortization, balloon and interest-only payment schedules, terms exceeding 30 years, alt-A or no documentation, and total points and fees that exceed 3 percent of the loan amount. In addition, a QM requires that the monthly debt-to-income ratio of borrowers not exceed 43 percent. For more on the ability to repay and QM standards under Regulation Z, see the Consumer Financial Protection Bureau's website at www.consumerfinance.gov/regulations/ability-to-repay-and-qualified-mortgage-standards-under-the-truth-in-lending-act-regulation-z. Return to text

7. This set of special questions has been included annually since the July 2011 survey. Return to text

8. In the SLOOS, questions about levels of standards define large and middle-market firms as firms with annual sales of $50 million or more, small firms as those with annual sales of less than $50 million, and very small firms as those with annual sales of less than $5 million. Return to text

9. Note that 14 of 22 foreign banks responded that they did not originate non-syndicated loans to small firms. Return to text

Back to Top