Changes in Consumer Credit
During the recent financial crisis, the Federal Reserve and other policymakers throughout the government took unprecedented actions to mitigate the fallout from severely distressed market conditions and support the flow of credit to consumers and businesses. Nonetheless, the level of credit outstanding for households has been very slow to rebound and remains lower than it was at the onset of the crisis. The reasons for the slow rebound are, without a doubt, complex and multidimensional. Still, it is worthwhile to examine the data and try to understand why credit growth is not more robust.

For this forum, I have chosen to focus my discussion on factors affecting the overall movements in credit card debt. The bulk of revolving credit in the United States today is held in the form of credit card debt. As the financial crisis developed in late 2008, the aggregate amount of credit card debt outstanding began to fall. Revolving credit has dropped every month since that time and is currently about 15 percent lower than it was at the time of the Lehman Brothers Holdings bankruptcy. Although our economy has experienced other long episodes in which revolving credit growth has slowed, we have never seen such a prolonged period of outright decline.

As overall consumer spending weakened significantly over the course of the recession and the early stages of the recovery, a proportionate decline in revolving credit used to finance purchases might actually have been expected. However, the decrease in revolving credit appeared to outpace the contemporaneous decline in spending during the recession, and, so far in the recovery, revolving credit has continued to decrease even as spending has turned up. This suggests that there are factors at work other than cyclical spending weakness. Within this context, it is helpful to consider the three main reasons that net borrowing–that is, the change in credit outstanding–can decrease: First, households can charge less on their revolving accounts; second, households can pay off a larger share of their balances each month; or third, households can default on (or lenders can charge off) their existing balances.

Taking the three factors in reverse order, consider first the role of cardholder defaults. As the economy weakened in 2008 and 2009, an increasing number of households found it difficult to pay their credit card bills on time. With nearly 10 percent of the workforce unemployed and many more underemployed, a significant number of households experienced sharply reduced incomes. Weakness in the housing market also contributed to financial strains, as many households could no longer easily tap into home equity to consolidate their card debt and lower their monthly payments.¹ In this adverse economic environment, it is perhaps not surprising that the charge-off rate on credit cards more than doubled from about 4 percent in 2007 to more than 9 percent in 2009. The rate of charge-offs has since declined from its peak but remains elevated. All told, we estimate that the rise in charge-offs can account for about one-third of the net decline in revolving credit growth from 2007 to 2009.

Another possible explanation for the decline in outstanding balances might be that households, in an effort to repair their balance sheets or bring down their debt burdens, have begun paying down their credit card balances faster than usual. But, on the whole, the data do not indicate that faster paydown is a significant factor. Historically, households tend to repay their credit card balances at a faster rate during good economic times and tend to slow this rate when economic activity is weak. And, over the past several years, this tendency appears to have held up. In 2006, the rate of credit card repayment was well above its long-run trend, probably reflecting strong incomes as well as ample home equity that could be tapped to pay off more expensive card debt. However, beginning in 2007, as housing markets weakened and unemployment climbed, households began to pay off their card debt at a significantly slower pace–a trend that extended into 2008 and 2009 as the economic downturn worsened. All told, the drop in the payoff rate has been more pronounced than in the recessions of 1990-91 and 2000-01. More recently, however, this trend has reversed, and as of August 2010, the repayment rate had risen to a more typical level. While this increase likely reflects a gradual improvement in the ability of cardholders to repay their debt, it could also be attributed to a shift in the composition of cardholders in bank portfolios toward more creditworthy borrowers as charged-off accounts were replaced with new accounts underwritten using stricter criteria. The bottom line is that accelerated payment rates on existing balances do not seem to have contributed importantly to the drop in credit card debt outstanding over the past couple of years.

Finally, consumers have been charging less on their credit cards. According to industry statistics, the amount of money charged on credit cards for purchases or cash advances fell around 10 percent between the third quarter of 2008 and the first quarter of 2010. This slowdown in new charges could be the result of a variety of factors. The significant overall drop in consumption during the recession no doubt cut into the demand for credit, as households simply opted to spend less than in the past. When they did spend, they may have been less willing to borrow to fund consumption given their experiences during the financial crisis, expectations for weaker economic conditions, and continued uncertainty about job prospects. Indeed, consumer preferences toward debt do appear to have shifted. Preliminary data from the 2007-09 Panel Survey of Consumer Finances (SCF) show a modest increase–from 35 percent in 2007 to more than 40 percent in 2009–in the share of households that believed that buying things on credit was a “bad idea.” Further, those households whose views about buying on credit became more negative between 2007 and 2009 reported reducing their charges substantially more than other households. Consumers also appear to be seeking less new credit: Applications for new credit accounts, as recorded in data from the national credit bureaus, remain significantly lower than were observed for most of the past decade.

Credit supply factors have also likely contributed to the decline in overall credit card outstanding balances. Households may have charged less because they had less credit available. In the SCF panel, about 44 percent of households with credit card debt in 2007 experienced a reduction in their credit limits by 2009. Data from the national credit bureaus indicate that credit lines peaked in the third quarter of 2008 and continued to fall over the course of 2009 and 2010. The average dollar value of combined credit lines available to cardholders fell from a high near $26,000 per cardholder in late 2008 to around $21,000 per cardholder by the third quarter of 2010, a decline of about 20 percent. Moreover, SCF data indicate that changes in credit limits are indeed related to credit card spending: Among households whose credit limit declined, SCF data show that the median amount of monthly new charges fell from $200 in 2007 to $50 in 2009, while among households whose credit limit did not decline, the median amount of new charges rose from $150 to $200. Although this relationship is not necessarily causal, credit line restrictions have likely played at least some role in the reduction in credit card borrowing.

Looking further at the supply side of credit card debt, card lenders did report retrenching during the financial crisis. According to the Federal Reserve Board’s quarterly Senior Loan Officer Opinion Survey on Bank Lending Practices (SLOOS), large fractions of banks tightened standards and terms on new and existing credit card accounts throughout 2008 and 2009.² In recent months, however, some banks reported having eased standards somewhat. In the most recent survey, published in October 2010, about two-thirds of banks thought that credit standards for prime borrowers on credit card loans and other consumer loans either were at their longer-run averages or would return to them over the next two years. In contrast, for nonprime borrowers, more than half of the respondents thought that standards would remain tighter through at least 2013 or would not return to longer-run norms for the foreseeable future.

In addition to reductions in existing credit lines, new credit card account solicitations also fell considerably during the recession. By early 2009, offers to households for new credit cards had dropped to around one-fifth of their count in 2006. Card solicitations have turned up over the course of 2010, but they remain well below their pre-crisis levels. In addition, consistent with the SLOOS, the data on credit card offers show that solicitations to borrowers with lower credit scores are rebounding more slowly than those to borrowers with higher scores.

Interest rates may have caused some households to reduce their credit card usage even if unused credit lines remained available. Although credit card interest rates declined in line with broader interest rates early in the financial crisis, card rates diverged from the broader rate environment by reversing this decline during 2009. Some of the rate increase likely reflects a rise in charge-offs, which increases card issuers’ costs of providing credit. However, the divergence from rates on other forms of credit that also experienced higher charge-offs indicates that a portion of the increase may have been in anticipation of regulatory changes, which I will discuss a bit later, that will restrict some card issuers’ ability to reprice credit.

Overall, then, the available data lead me to conclude that, in large part, the decline in revolving consumer credit outstanding is due to a combination of higher charge-offs, tighter credit, and less consumer willingness to take on debt, but probably not to widespread increases in discretionary paydowns of existing debt.

Although households account for the vast majority of credit card loans and credit card spending in our economy, the market for small business credit cards has grown considerably over the past 10 to 15 years. After checking accounts, credit cards are the second-most-common financial product used by small businesses. Small business cards are structured to cater to business needs with features, pricing, and underwriting unique to their typical usage. Issuers provide several services specifically for small businesses, such as employee cards with customizable spending limits and detailed spending statements each month or quarter. Also, small business cards often have higher credit limits than personal cards to facilitate the higher spending needs of small businesses.

Small businesses are noticeably less likely than households to carry a balance on their cards. As of the end of 2009, 83 percent of small businesses used credit cards. Of those using credit cards, 64 percent used small business cards and 41 percent used personal cards. Despite the widespread use of credit cards, only a minority of small businesses–18 percent–reported borrowing on credit cards. In comparison, nearly one-half of households reported carrying a balance on their credit cards.³ Thus, although most small businesses appear to use credit cards for transactions purposes, and perhaps as a source of short-term credit, the data suggest that only a small fraction of them rely on credit cards as a source of longer-term credit. Yet even if firms do not carry a balance, reductions in the size of their credit card lines may strain their cash flow and force them to cut spending or require them to use more expensive forms of short-term credit, such as trade finance.

Thursday’s bond market has opened in positive territory following much weaker than expected consumer spending news. The stock markets are showing minor gains with the Dow up 27 points and the Nasdaq up 10 points. The bond market is currently up 15/32, which will likely improve this morning’s mortgage rates by approximately .125 of a discount point. Preventing a slightly larger improvement in rates was weakness late yesterday after the FOMC meeting.

The Commerce Department announced this morning that retail level sales fell 0.1% last month. This was well off forecasts of a 0.7% increase, meaning that consumers were spending much less than expected. Even if volatile auto-related sales are excluded, sales fell much more than expected. This is very good news for the bond market and mortgage rates because consumer spending makes up two-thirds of the U.S. economy. If consumer spending is still falling, the broader economic recovery cannot be close. Generally speaking, a weak economy is a better environment for bonds and makes mortgage-related bonds more attractive to investors.

Also posted this morning were weekly unemployment figures from the Labor Department. They reported that 558,000 new claims for benefits were filed last week. This was an increase from the previous week, but more importantly, analysts were expecting to see a decline in new claims. However, since this data basically tracks only a week’s worth of claims, it usually has little impact on mortgage rates and has not influenced trading this morning.

Early this afternoon we will get the results of today’s 30-year Bond auction. This sale is not as important to mortgage rates as yesterday’s 10-year sale was. But if the auction is met with an overly strong demand from investors or a particularly weak interest, we may see bond prices move enough during afternoon trading to cause revisions to mortgage rates. The results will be posted at 1:00 PM ET.

Tomorrow morning brings us the release of three reports. The first is July’s Consumer Price Index (CPI) at 8:30 AM. The CPI is one of the most important reports we see each month. It measures inflation at the consumer level of the economy. There are two readings in the report- the overall index and the core data reading. The more important of the two is the core data because it excludes more volatile food and energy prices. Current forecasts call for no change in the overall index and a 0.1% increase in the core data reading. Declines in the readings, especially in the core data, should lead to a bond rally and lower mortgage rates. However, stronger than expected readings will likely cause a spike in mortgage pricing tomorrow.

The remaining two pieces of data are relevant to mortgage rates but not nearly important as the CPI is. The second report of the day is Industrial Production data for July. This report gives us a measurement of manufacturing sector strength by tracking output at U.S. factories, mines and utilities. It is considered to be of moderately high importance and may cause movement in mortgage rates. Analysts are currently expecting to see a 0.4% increase in production between June and July. A larger increase in output could lead to higher mortgage rates tomorrow, but only if the CPI’s results are a non-factor in rates.

The last report of the day will come from the University of Michigan who will release its Index of Consumer Sentiment for August at 9:45 AM. This index gives us a measurement of consumer willingness to spend. If confidence is rising, then consumers are more apt to make large purchases. This helps fuel consumer spending and economic growth. A drop in confidence will probably help boost bond prices. If the index rises, indicating that confidence is rising and spending is likely to continue, we may see mortgage rates move higher Friday morning. However, this is the least important of the day’s three reports and will probably have the least impact on rates.

If I were considering financing/refinancing a home, I would….
Lock if my closing was taking place within 7 days…
Lock if my closing was taking place between 8 and 20 days…
Float if my closing was taking place between 21 and 60 days…
Float if my closing was taking place over 60 days from now…

This is only my opinion of what I would do if I were financing a home. It is only an opinion and cannot be guaranteed to be in the best interest of all/any other borrowers.

Rate Lock Advisory – Wednesday Nov. 26th

Wednesday’s bond market has opened in positive territory following weaker than expected economic news. The stock markets are currently in positive ground after initially opening with losses. The Dow is now showing a 42 point gain while the Nasdaq is up 28 points. The bond market is currently up 19/32, which should improve this morning’s mortgage rates by approximately .250 of a discount point.

The first piece of data released this morning was October’s Durable Goods Orders that showed a drop of 6.2% in new orders and revised September’s orders lower than previously announced. Analysts were expecting to see a 2% drop in October’s orders, meaning that demand for big-ticket products was much weaker than thought. In fact, this was the largest monthly decline in approximately two years. That is good news for bonds and mortgage rates, because the slowing economic activity makes mortgage related bonds more attractive to investors.

The second was Oct ober’s Personal Income and Outlays data, which gave us mixed results. The bad news came in the income portion of the report that revealed a 0.3% rise in income compared to forecasts of a 0.1% increase. This means that consumers have more money available to spend than was expected. However, the good news was that they spent less than analysts had predicted. What was supposed to be a 0.7% decline in spending actually came in at a 1.0% drop. With consumer spending making up two-thirds of the U.S. economy, the weaker than expected spending is taken as good news for bonds.

This month’s revision of the University of Michigan’s Index of Consumer Sentiment was also favorable to bonds and mortgage rates with a reading of 55.3. This was much lower than the 58.0 that was expected, indicating that consumers were less optimistic about their own financial situations than analysts had thought. This means they are less likely to make large purchases in the near future.

The last report of the day was October’s New Home Sales figures that showed that sales of newly constructed homes fell to its lowest level in almost 18 years. While this is generally good news for bonds and mortgage rates, this data is not considered to be oh high importance to the markets, therefore, its impact ton today’s trading and mortgage rates has been minimal.

The bond market will close at 2:00 PM ET today ahead of tomorrow’s Thanksgiving Day holiday and will reopen Friday morning. There is no relevant data scheduled for release Friday, so I am expecting to see a very light and thinly traded day. The bond market will also close at 2:00 PM Friday, so look for little activity that day.

If I were considering financing/refinancing a home, I would…. Lock if my closing was taking place within 7 days… Lock if my closing was taking place between 8 and 20 days… Float if my closing was taking place between 21 and 60 days… Float if my clos ing was taking place over 60 days from now… This is only my opinion of what I would do if I were financing a home. It is only an opinion and cannot be guaranteed to be in the best interest of all/any other borrowers.

Rate Lock Advisory – Thursday Nov. 13th

Thursday’s bond market has opened in negative territory, erasing part of yesterday’s late rally that came as a result of strong stock losses. The stock markets have opened in negative ground, continuing yesterday’s selling. The Dow is currently down 90 points while the Nasdaq has lost 27 points. The bond market is currently down 4/32, but we will still likely see a small improvement in this morning’s mortgage rates of approximately .125 of a discount point due to strength in bonds late yesterday.

This morning’s first piece of news was the release of September’s Goods and Services Trade Balance report. It gave us the size of the U.S. Trade Deficit, showing a $56.5 billion deficit. That was a little smaller than forecasts of $57.0 billion, but this data is not considered to be of high importance to the markets and has had little impact on this morning’s trading or mortgage pricing.

The other news released this morning was weekly unemployment figur es from the Labor Department. They reported that new claims for benefits jumped to 516,000 last week, exceeding forecasts of 479,000. The previous week’s figures were revised to 484,000, meaning analysts were expecting to see a small decline in claims when we actually saw a sizable jump. While this data is not considered to be of high importance because it tracks only a week’s worth of filings, it can influence trading and rates when it varies from forecasts such as today’s variance.

There are two reports scheduled for release tomorrow morning with one of them considered to be very important to the markets. October’s Retail Sales report is the first and the highly important one because it measures consumer spending. Since consumer spending makes up two-thirds of the U.S. economy, any related data is watched closely. If this report reveals weaker than expected sales, the bond market should thrive and mortgage rates will fall. Current forecasts are calling for a drop in sales of approximately 2.1%.

The second report comes late tomorrow morning when November’s preliminary reading of the University of Michigan’s Index of Consumer Sentiment will be released. This index measures consumer confidence, which gives us an indication of consumer willingness to spend. It is expected to show a reading of 57.0, down from October’s final reading of 57.6.

If I were considering financing/refinancing a home, I would…. Lock if my closing was taking place within 7 days… Lock if my closing was taking place between 8 and 20 days… Float if my closing was taking place between 21 and 60 days… Float if my closing was taking place over 60 days from now… This is only my opinion of what I would do if I were financing a home. It is only an opinion and cannot be guaranteed to be in the best interest of all/any other borrowers.