Forget Whatever You Knew About Consumer Credit Since 2006

Written by Steven Hansen

Sit back, grab a coffee – and forget whatever you knew about consumer credit since 2006.  The Federal Reserve literally changed the baseline in April 2012 (released today) likely due to double counting of certain types of credit.

The market expected consumer credit to expand $10.0 to $12.7 billion versus the reported $6.5 billion.  What the market did not expect was a significantly revised data base.

What the simple headline said:

Consumer credit increased at an annual rate of 3 percent in April. Revolving credit decreased at an annual rate of 4-3/4 percent, while nonrevolving credit increased at an annual rate of 7 percent.

Econintersect does not believe the seasonal adjustment methods used in the headlines are accurately conveying the situation for a variety of reasons.   To begin, it does not explain the new baseline which was warned in a April 2012 Fed Press Release:

The Federal Reserve Board on Monday announced that it has restructured the G.19 statistical release, Consumer Credit, to reflect regulatory filing changes for U.S.-chartered depository institutions and, in addition to the data currently reported on level of credit outstanding, the release will now report data on the flow of credit. The revised data will be made available with the release of the April report on Thursday, June 7.

Savings institutions now file the same regulatory report as U.S.-chartered commercial banks. The U.S.-chartered commercial banks sector and the savings institution sector, which were previously shown separately, have been combined into a new sector called depository institutions. The previously published series for U.S.-chartered commercial banks and savings institutions will continue to be available as separate series in the Federal Reserve’s Data Download Program (DDP).

The new flow data represent changes in the level of credit due to economic and financial activity, rather than breaks in the data series due to changes in methodology, source data, and other technical aspects of the estimation that affect the level of credit. Access to flow data allows users to calculate a growth rate for consumer credit that excludes such breaks.

These changes will be accompanied by revisions to the estimates of outstanding consumer credit back to January 2006 and reflect improvements in methodology and a comprehensive review of the source data.

The Fed was not kidding.  This data series has been significantly revised.  Was the Fed double counting student loans at times?  The credit flow data is interesting, and now allows an easier production of graphs such as the one below:

Econintersect spends time on this generally ignored data series as the USA is a consumer driven economy. One New Normal phenomenon is the consumer shift from a credit towards a cash society – a quantum shift which changes the amount of consumption. Watching consumer credit provides confirmation that this New Normal shift continues.

The Econintersect summary of the data based on unadjusted data:

Year-over-Year Growth Rate Change in rate of growth from Previous Month Trend
Total Credit 4.9% 1.6% 15 months with positive YoY growth
Revolving credit 1.4% 0.3% sixth month of YoY growth after 33 months of decline
Non-revolving credit 6.6% 0.1% 20 months with positive YoY growth

What year-over-year change looked like last month:

What the year-over-year change looks like now:

If student loans are backed out, consumer credit grew 0.4% year-over-year, and grew 0.1% month-over-month.

Graph from last month:

This month’s graph:

Econintersect backs out student loans as they are currently consuming an unusual and inordinate portion of USA consumer loans.

Graph from last month:

Graph from this month:

Note: Student Loans have never declined during this period. Where student loans are shown at 100% of the growth, consumer credit (including student loans) could have declined in that particular month even though student loans outstanding increased.

A good background article was written by Frederick Sheehan.

The Federal Reserve reports credit divided between revolving and non-revolving. The majority of revolving credit is from credit cards, while non-revolving credit includes automobile loans, student loans, and all other loans not included in revolving credit, such as loans for mobile homes, education, boats, trailers, or vacations.

Graph from last month:

Graph from this month:

Caveats on the Use of Consumer Credit

This data series does not include mortgages, and is not inflation adjusted.

The graph below shows consumer credit outstanding (this data series does not include mortgages) is slightly less than 23% of annualized consumer spending – down from a high of over 26% in the 2000s, but still above the averages before the mid 1990s. [note: Fred has not updated their graphs – this is based on March 2012 using the old data set]

To get a feel of inflation adjusted consumer credit, the following graph is inflation adjusted consumer credit using the CPI-U (less shelter) – this is expressing consumer credit in 1982 dollars. It is evident on an inflation adjusted basis, consumer credit is beginning to grow.  [note: Fred has not updated their graphs – this is based on March 2012 using the old data set]

Econintersect believes consumer credit levels are now in its historical channel from the 1990’s.

Related Articles

All Consumer Credit Posts