There has been much speculation in recent reports about the rise in interest rates for new credit card offers. The most common conclusion is that this increase is attributable to the Credit CARD Act (CARD Act), legislation regulating the credit card industry that became effective in February 2010.
CardHub.com conducted a study to determine whether or not this hypothesis is true. The key question of the study, however, was not whether or not credit card interest rates have gone up – because they have – but whether or not the interest that consumers are actually paying on credit card debt has gone up as a result of CARD Act.
To evaluate the CARD Act’s impact on the cost of consumer credit card debt we followed three different methodologies: one based on intuitive analysis of the legislation and current economic environment, another based on an analytical analysis of historical data for interest rates and debt, and lastly an analytical analysis based on a statistical model.
To view the data on historical interest rates used in this study, please use the following link:
Analysis based on intuition:
The impetus of this study was that intuitively it did not make sense for the CARD Act to cause a rise in the cost of consumer credit card debt. This is because the CARD Act did not aim at stopping credit card companies from charging consumers whatever they want in interest – they were allowed to do it before and are still allowed to do it now – but instead it aimed at preventing the “bait and switch” tactic that many credit card companies used.
Previously, credit card companies could offer a lower interest rate on their promotional materials to lure a customer in, and then suddenly increase the interest rate on their existing balance without reason. The customer was suddenly stuck trying to pay down their balance at a much higher interest rate than they signed up for. In other words, the interest rate on the credit card company’s promotional materials did not match the interest rate the consumer actually ended up paying.
Now, the CARD Act requires that the interest rate on the promotional materials match the interest rate that the consumer actually ends up paying. It did not change what the credit card companies could charge, merely the transparency with which they had to disclose this information.
Furthermore, the CARD Act does not prevent credit card companies from increasing the interest rate on future transactions, it only stipulates that they must give the consumer 45 days notice, as opposed to the previous industry standard of 20 days notice. This again points to the fact that the companies do not have to lose any revenue from interest charges, just that they have to be more up front about them.
Another element of the CARD Act that people point to is its regulation on fees. The common logic is that given credit card companies lost revenue in fees, they are making up for it by increasing interest rates. However, the lion share of penalty fees comes from consumers with subprime credit (i.e. FICO credit score below 660), as this group of consumers is more often hit with fees for going over their credit limit or missing payments. However, not all credit card companies offer credit to subprime credit customers and therefore some (e.g. American Express) were not impacted by the CARD Act fee regulations in any significant way.
Since consumers with subprime credit have low credit limits (usually between $300 and $1000), they are not the segment of consumers that generates the bulk of the revenue in interest for credit card companies. Instead, it is consumers with prime or superprime credit and high credit limits that generate the most revenue in interest. Therefore, it would not make sense for companies who offer credit to both people with subprime and superprime credit to subsidize their subprime business by raising interest rates on prime/superprime credit customers because it would make these companies less competitive with those who are only in the prime/superprime market.
Analytical analysis of historical data for interest rates and debt:
The next methodology in examining the CARD Act’s impact on the cost of credit card debt was to compare historical data on interest rates and debt from 1990 to 2010. In order to normalize the data, we subtracted the Prime Rate from the average interest rate for credit cards for each month we evaluated. It was this margin between the average interest rate and the Prime Rate that we used to compare what banks were actually charging consumers above the Prime Rate – which is the portion of the interest rate they have control over.
Overall Historical Data Analysis
The highest margin that we have seen since the CARD Act became effective was in February 2010, the month the CARD Act was implemented, with a margin of 11.01 %. Since then, the number has consistently fallen. Even with a margin of 11.01%, this is still lower than the margins we saw in the early 1990s, when the margin was as high as 11.68% in August of 1992. This was during a recession that was much less severe, with unemployment at 7.6% in August 1992 compared to 9.7% in February 2010, and delinquencies at 5.04% compared to 5.88% in February 2010.
The fact that there were higher margins at a time when there were lower delinquencies and unemployment rates illustrate the fact that it is the economic environment, and not the CARD Act, that has been affecting these margins.
Recent Historical Data Analysis
Although the above analysis of the overall historical data could stand alone in proving that it is the economic environment driving these margins, we also took a closer look at recent months to further illustrate this point.
If you look at the data from August of 2008 on, you will see that the margin jumps from 6.94% in August of 2008 to 8.03% in November of 2008. This is a huge jump of more than 1% (a 15.7% increase) during a time in which the CARD Act had not even been signed into law (the CARD Act was signed into law in May 2009). You will also notice that during this time delinquencies went from 4.83% in August 2008 to 5.72% in November 2008. Since the CARD Act was obviously not a factor, we can attribute this large jump in the margin to the same large jump in delinquencies during this time.
Further strengthening the link between the rise in delinquencies and the rise in the margin, you can see from the data that as delinquencies jumped almost another 1%, from 5.72% in November 2008 to 6.61% (the highest it’s been in the last 20 years) in February 2009, the margin jumped by another 1.69% to 9.72%.
Since the beginning of the recession, we have seen the margin go from 5.25% in November of 2007 to its peak of 11.01% in February of 2010. It’s clear from the data that almost 78% of this increase occurred prior to the CARD Act – from 5.25% in November 2007 to 9.72% in February 2009 – and is therefore explained by the rise in delinquencies, and not the CARD Act.
The remaining 22% increase from a margin of 9.72% in February of 2009 to 11.01% in February 2010, is less obviously correlated to delinquencies, but we believe is still a result of the economic environment. Although delinquencies did not continue to rise during this time period, they remained at an unprecedented level – well above 6% – for almost an entire year. Delinquency rates that remain at unprecedented high levels are just as alarming to banks as delinquency rates that consistently rise. This time period also saw higher levels of unemployment than we’ve seen in the last 20 years, with the unemployment rate reaching its peak of 9.9% in November 2009. Both of these factors created an unprecedented default risk for banks, which made them want to further increase interest rates in order to better protect themselves.
These are strong indications that it was the economy, and not the CARD Act, driving the margins. Additionally, we now see that the margin was already back down to 10.19% in November of 2010, which is very close to the 9.72% it reached in February of 2009 before the CARD Act was signed into law.
Analytical analysis based on statistical model:
Our last method for examining the CARD Act’s impact on credit card interest rates was based on a statistical model. We built a statistical model for the time period between May 1991 to November 2008, using data such as the Prime Rate, delinquency rates, charge-off rates, and unemployment rates, to predict what the interest margins should be. The model we built explained 84% of the observed changes in the interest margin during the time period for which it was built. When we applied this model, based on underlying economic factors, to the time period between February of 2009 and February of 2010, the statistical model predicted much higher margins than what we have actually observed. This further indicates that it is the economy, and not the CARD Act, that is affecting credit card interest rates.