Credit Conditions Surveys Methodology
Published January 1, 2009 | January 2009 issue
This special focus on credit conditions in the Ninth District was done through a series of four polls, two of which had special questions included in regularly scheduled annual surveys, and two of which were conducted specifically for this issue of the fedgazette.
Every year, the Federal Reserve Bank of Minneapolis conducts so-called business-conditions and manufacturers’ polls as part of its annual report on economic conditions and forecast for the coming year in the Ninth District.
The business-conditions poll is sent to a network of business contacts maintained by the Minneapolis Fed. The poll asks a total of 31 questions, including a small set of special questions designed to investigate a relevant economic issue. This year, a total of four questions focused on credit. The poll was sent out in late October, and responses were accepted through mid-November. A total of 334 responses were received from an original mailing of 1,060 surveys.
The Minneapolis Fed also conducts an annual poll of manufacturers in cooperation with the Minnesota Department of Employment and Economic Development. Surveys of Minnesota manufacturers were produced and processed by the Minnesota DEED. The Federal Reserve Bank of Minneapolis conducted the survey for manufacturers for all other parts of the Ninth District.
This year the poll included four special questions on credit conditions. From the approximately 17,000 Ninth District manufacturers, a stratified random sample of 2,698 businesses was drawn and stratified based on employment size; the sample includes 100 percent of establishments with more than 49 employees, 20 percent with five to 49 employees and 3 percent with fewer than five employees.
A postcard survey was mailed to each of the selected businesses. The mailing was sent in late October/early November, and a second mailing was sent in mid-November to Minnesota businesses that did not respond to the first mailing. A cutoff for survey responses was Dec. 1. A total of 446 usable surveys were received, for a response rate of 15 percent. Of the usable surveys, 76 were from manufacturing respondents indicating that their specific location had no on-site manufacturing.
Survey results were tabulated for all manufacturers. The confidence interval for sampling error was calculated. The 95 percent confidence interval for this poll’s results is plus or minus 4.6 percentage points. Results are also subject to errors introduced by other factors, such as the wording of questions and differences between survey respondents and nonrespondents.
The fedgazette also conducted two special surveys on credit conditions: One partnered with state banking and credit union associations; the second partnered with state chambers of commerce. State associations representing banks and credit unions in all or parts of six district states agreed to electronically distribute an eight-question poll to members for which they had contact information. A total of 12 associations participated, with representation of at least two associations in each district state and one in the Upper Peninsula of Michigan.
Combined, the poll was sent to a total of about 6,100 contacts, which included an unknown number of duplicate contacts from organizations that were members of more than one of the associations involved. The poll was conducted from Nov. 4 to Nov. 12. Members had the option of responding via e-mail or fax, or taking the survey online. A total of 367 responses were received. Only one response was received from the U.P., which prevented the fedgazette from breaking out U.P. results separately.
The fedgazette also organized a poll of businesses through the state chambers of commerce from Minnesota, Montana, North and South Dakota and Wisconsin. These organizations electronically forwarded a 10-question poll to members. The survey was conducted from Nov. 5 to Nov. 14. A total of 8,263 surveys were originally distributed by state chambers to their memberships. Members had the option of responding via e-mail or fax, or taking the survey online. A total of 1,722 surveys were received.
The state chambers also sent the survey to local affiliate chambers of commerce, who were encouraged to forward the poll to their local members. Anecdotal evidence suggests that a few local chambers forwarded the survey to their local membership. The Kalispell (Mont.) Chamber of Commerce, for example, distributed the survey to those attending a conference on Nov. 12 and received 60 usable responses. The poll was also sent to a handful of the largest chambers of commerce in the U.P., but no responses were received.
Unlike the manufacturing survey, the sample populations for the other three surveys were not randomly drawn. Therefore, results for these polls cannot claim to be representative.
Partner organizations for this special fedgazette focus on credit include: Michigan Bankers Association, Minnesota Bankers Association, Montana Bankers Association, North Dakota Bankers Association, South Dakota Bankers Association; Independent Community Bankers of Minnesota, Montana Independent Bankers Association, Independent Community Banks of North Dakota, Community Bankers of Wisconsin, Minnesota Credit Union Network, Montana Credit Union Network, Mid-America Credit Union Association; Minnesota Chamber of Commerce, Montana Chamber of Commerce, North Dakota Chamber of Commerce, South Dakota Chamber of Commerce and Industry, Wisconsin Chamber of Commerce Foundation, Minnesota Department of Employment and Economic Development.
Survey results data. [xls]