Worst Practices: Residual Interest
Professor LoPucki's APR issue might not be two-cycle billing as many of the commentators think. Just as likely, it is residual interest (a/k/a trailing interest), the often ignored, but just as potent cousin of double-cycle billing.
Residual interest has not gotten nearly as much attention as it should (and it is often confused with double-cycle billing). Residual interest is a nasty billing trap that drains away discretionary income (also known as potential savings) from American consumers, and should be at the top of the Congressional hit list for predatory credit card billing practices.
Residual interest is not an intuitive concept. The key to understanding it to remember that finance charges apply to the average daily balance outstanding during a billing cycle. Billing cycles start, however, before the old billing cycle's bill has been sent out and payment has been received. If payment on the preceding cycle is received in full and on time, then it is not included in the average daily balance. But if even a penny of it is late or not paid, then the previous cycle's balance is treated as outstanding for the days that would otherwise be in the grace period.
By way of comparison, double-cycle billing calculates an average daily balance based on the current and past billing cycle (this can actually be beneficial to cardholders in some circumstances, such as when there was a zero balance in one cycle). Double-cycle billing can be combined with residual interest, so that charges from a third billing cycle that are paid off during the grace period still are figured into the balance calculation. What both residual interest and double-cycle billing share is that they calculate the average daily balance to which the finance charge is applied by including amounts that have been timely repaid.
To illustrate how residual interest works: if you charge $500 in cycle 1 which ends on September 30, you do not get a bill until, say, October 7, after the start of the October billing cycle. You then pay $450, which the issuer receives on October 12 (on time). You then on October 16 charge another $200. Your total daily balance for the October billing cycle (closing on October 31st), is calculated thus:
Oct. 1-Oct. 12 (12 days @ $500)=$6,000
Oct. 13-Oct. 15 (3 days @ $50)=$150
Oct. 16-Oct. 31 (16 days @$250)=$4,000
Total Daily Balance: $10,150
This is then divided by the number of days in the cycle (31), yielding an average daily balance of $327.42, to which the finance charge is applied. 14.49% compounded daily for 31 days on $327.42 is $4.05.
Notice that you paid $450 back on time, but are still assessed interest on it. This would be true if you had repaid all but a penny on time. But for residual interest, the average daily balance would be ($250*16 + $50*12)/31=$161.29, so interest would be applied to a balance that is 55% lower! At 14.49% APR interest, compounded daily, the difference is that between paying $4.05 in interest for the month for residual interest or $2.02 (no residual interest).
That doesn't look like a big deal (although there is a lot to be said for a business model that makes lots of money in small increments), but consider if the balances had an extra zero, and were carried for a year, instead of a month. In such a case, residual interest would be the difference between paying $510.42 in interest instead of $255.19 in interest. Again, the extra $255.23 doesn't seem like a big deal...unless you consider it in the context of many families' discretionary income.
For a household making less than $50,000 per year, this extra $255.23 in interest represents 13% of their discretionary income. Even for wealthier households, earning $50,000-$99,999 per year, it is over 2% of discretionary income. Imagine if every household's discretionary income rose 2%--that would be serious economic growth. Billing tricks and traps like residual interest drain away spending power from American consumers.
Here's the other thing about residual interest: it's fully disclosed. This tells us that TILA disclosures are useless. The published APR is 14.49%, but it is applied in a way that no one understands to yield an effective APR of 29.23% on the balance that was actually revolved. Grace periods only apply if the balance is paid in full and on time. Otherwise, residual interest kicks in. If Lynn LoPucki, one of the leading credit scholars in the world, didn't catch the residual interest trap (or the double-cycle billing, which is almost certainly in addition to residual interest) in his card disclosure, can we really expect that any one else will? TILA has become a means for obfuscating credit prices, not for their disclosure. If we want credit markets to work and to enable consumer responsibility, we have to fix TILA, and that starts by banning billing practices that manipulate disclosure rules.
[Updated with technical corrections, 10.25.08 at 8:16pm. Updated again with further technical corrections 10.26.08 at 3:50pm.]
First off, there is no September 31. :) So I gather the difference is about the period from Oct. 1 to Oct. 12 (when $450 payment was received). Should the interest charge for that period be based on $500 or $50? Is that the crux of the question?
The answer is $500, because the borrower has that $500 until Oct. 12 when he/she reduced the outstanding to $50. When the borrower has the use of that money, I don't see why it's unfair to charge interest on it. Had the payment been $500, no interest would be charged. That's an exception to the general rule "if you borrow money, you pay interest" which is not that hard to understand. I think in the UK it's specifically stated that way. For each day the money is outstanding, interest accrues. If you pay the statement balance in full, the interest is then credited back.
Moral of the story: always pay in full. If you are not going to pay in full, pay early. You don't have to wait for the statement. You can pay at any time, even before the statement period is over.
Posted by: TFB | October 25, 2008 at 04:20 PM
September 30th it is.
TFB gets to the heart of the issue. If you pay in full, then there is a grace period for timely repayment before interest accrues; if you pay even on cent less than the full balance, however, there is no grace period.
It wouldn't be unfair to charge interest on the full $500, rather than just $50 if there were no grace period. But federal law requires a grace period of 14 days (arguably it can be read as 15 days--the Fed/OTS/NCUA regs would add a week to that). So consumers have a default expectation of a grace period, and the logical assumption about a grace period is that funds repaid within that period do not accrue interest, not that funds repaid within that period do not accrue interest if any only if there is full repayment of the previous cycle's balance. To have the grace period depend on as little as one penny makes no sense other than as a way to trap the unwary.
Despite my previous questioning of behavioral law and economics, framing and default rules do matter. To indulge in a football analogy, while they just aren't going to carry the ball past the goal post in most cases, they are good for some yardage. If the default rule were stated as (1) interest accrues as a default, and (2) you get a refund if you pay in full by date X, I suspect that the framing would make consumers less likely to use credit cards--which is why the card industry wants to maintain status quo.
Posted by: Adam Levitin | October 25, 2008 at 06:29 PM
The daily balance computation is off: $250*16 + $50*20 + $500*12 = $11,000, not $10,000. The $50*20 appears to be spurious (i.e. $10,000 should be the correct answer).
Posted by: Anonymous | October 25, 2008 at 07:02 PM
Thank you. Correction noted. Two gaffes and counting on this one.
Posted by: Adam Levitin | October 25, 2008 at 07:21 PM
Adam, a grace period is not required by federal law, correct? The rule is that if a creditor does provide for a grace period, it must be at least 14 days. Correct?
Posted by: Anonymous | October 25, 2008 at 09:48 PM
The grace period is required. The grace period is a result of federal regs, not federal statute, and is not actually presented as a grace period, per se. The relevant regs are in Reg Z under the Truth in Lending Act.
12 C.F.R. 226.5(b)(2)(ii) provides that
"The creditor shall mail or deliver the periodic statement at least 14 days prior to any date or the end of any time period required to be disclosed under Sec. 226.7(j) in order for the consumer to avoid an additional finance or other charge.10 A creditor that fails to meet this requirement shall not collect any finance or other charge imposed as a result of such failure."
226.7(j) provides:
"Free-ride period. The date by which or the time period within which the new balance or any portion of the new balance must be paid to avoid additional finance charges. If such a time period is provided, a creditor may, at its option and without disclosure, impose no finance charge when payment is received after the time period's expiration."
Even if there is no free-ride period under 226.7, 226.5 creates a 14 day one because the statement has to be sent out at least 14 days before payment is due. The bill can't be sent out until the cycle is over, so there has to be 14 days minimum before interest can be collected. (Of course, a cycle could be shortened...)
It's a good question whether a "no interest, but only if paid in full" interpretation is consistent with 226.5, much less the Truth in Lending Act's requirement that a grace period be clearly stated (including if there is not one).
Posted by: Adam Levitin | October 25, 2008 at 10:03 PM
From an examination of my bill, it seems that Alan White's explanation is the correct one. My balance at the end of the August cycle was $9,607.61. The calculation of finance charge shown on the statement ($122.22) applies a "nominal" APR of 14.9% to a "Balance Subject to Finance Charge" of $9,620.44. $9,620.44 multiplied by .149 and divided by 12 is $119.45. The statement explains that the 14.9% is compounded daily, thus perhaps accounting for the remaining $2.77. So I was not charged interest on an actual extension of credit. I was charged interest on an imaginary one -- one imagined because my balance once went almost that high. Nor was I charged an APR, I was charged a compounded APR (if there can be such a thing). This can't be the Truth in Lending law that was initially enacted. What happened during the 30 years I wasn't paying attention?
Posted by: Lynn M. LoPucki | October 26, 2008 at 09:51 AM
Still not sure about that first daily balance calculation. Seems to me that (under residual interest) the balance was
- $500 for October 1 - 12 (12 days)
- $50 for October 13 - 15 (3 days)
- $250 for October 16 - 31 (16 days)
giving an average daily balance of ($250*16 + $ 50*3 + $500*12)=$10,150. I think that in
"($250*16 + $ 50*20 + $500*12)=$11,000"
there are two problems: the $50 balance is over-counted (because it's included again in the $250 balance after the $200 purchase) and October 12 is double-counted (included in $50 * 20 as well as $500 * 12--notice that 20 + 12 = 32 :))
Posted by: Christopher Hundt | October 26, 2008 at 02:11 PM
Three gaffes, and I'm out. I'm grateful for Christ Hundt, Anonymous, and TFB catching my arithmetic and calendar mistakes and have updated the post to reflect the correct calculations. Let's hope that's it for the snafus. It takes me back to the SAT, when I aced all the algebra and geometry, etc, but blew the 2+2 arithmetic problems (=5?).
My arithmetic and calendar mistakes, however embarrassing, do not undercut the argument about the invidious nature of residual interest. Instead, they point out just how difficult it is to calculate the cost of revolving a balance. Despite these mistakes, there are many folks who are more arithmetically challenged than I--calculating what the cost of revolving a balance is just not realistic for even diligent consumers. To do so, a consumer needs to know (1) what the average daily balance is, which requires knowing (2) how it is defined (single or double-cycle and that residual interest is included) and (3) when charges and credits will be applied to the account, which means knowing processing and US mail speeds, and (4) how to compound the balance daily, which requires knowing (5) if the compounding is based on a 360 or 365 day year (I have assumed a 365 day year for my calculations). I would posit that there is no one in the country who can or does know these pieces of information when they revolve a balance, and even if they were all knowable, it likely would not be worth the time to calculate, because that would only tell you the cost for a specific card, not which card is the cheapest. And this isn't including any fees, etc.
Posted by: Adam Levitin | October 26, 2008 at 03:06 PM
Adam - welcome to the real world. Consider the average reading and comprehension level of adults in the US is roughly the fifth grade. You can quickly see why the lending industry is so determined to obfuscate.
Take a quick guess at how many people are capable of figuring out their REAL debt situation with a fundamental tool such as a spreadsheet.
Then consider the expertise and technology we are up against.
I'm guessing that portions of 80+% of the population are little more than something to be harvested.
The emails we get from people in trouble indicate a profound lack of basic legal and financial understanding. It is almost impossible to imagine how much the average person has "donated" to the lending industry as a result of fundamental ignorance.
Posted by: Dave Mortensen | October 26, 2008 at 09:01 PM
Perhaps you are looking at this the wrong way.
Assume my credit cycle is from the 1st to the 30th of the month and I have a 21 day grace period. If I buy something for $100 on January 1 and pay for it on February 21, I have just received a FREE 52 day loan. (Yeah, the credit card company still gets some money, which I ultimately pay, so it is not really free).
If I do not pay it all on February 21, but wait until March 21, I have to pay some interest, but I still get the benefit of using the card, and even got a free 30 day loan (I think).
The thing is, the grace period is different from most other forgiveness program. It is automatic.
Perhaps we should have credit card companies assess all interest and note that the interest will be forgiven if the balance is paid in full. But that will not work, because some idiots will pay the entire balance due and some enterprising lawyers will file a class action suit.
Sometimes finance is complicated. I think that residual interest is not insidious and is a good way to go.
Posted by: Allan | October 26, 2008 at 10:16 PM
To oversimplify, let's say half the credit card users pay their credit card bills in full every month and never pay interest (Group A). The other half never pay in full and always carry a balance and pay interest (Group B). Group A cares not a whit about the nuances of how interest is calculated since they never pay any. Group B should care but realistically don't because they are unsophisticated, under financial pressure, and probably trying to figure out how to get more access to credit rather than seriously taking action to get out of the revolving credit trap they are in. Arguably,more or better disclosure, meaningless to Group A, will not help those in Group B since they are not making reasoned financial decisions on the basis of variations in APR. TILA's primary function is to highlight those differences so that borrowers can make informed choices between credit products, not to warn borrowers away from credit altogether. Amending TILA and Reg. Z will not likely help Group B. Getting Group B off the revolving credit addiction will require a cultural change --- something that is extremely difficult to legislate.
Posted by: Thomas Perkins | October 27, 2008 at 08:53 AM
Allan is correct than when you pay in full and on time you are (ignoring annual fees and interchange and any other transaction fees, none of which are actually negligible) you can have a 52 day interest free loan (or whatever it might be with the particular grace period). When seen in this light, it is card issuers in their good graces granting an interest free loan, and residual interest is just a small paring back of this magnanimity.
But to call this interest free isn't quite right, either--the grace period cost of funds to the issuer is figured into the interest rate (and other fees). The longer the grace period, the higher the interest rate the issuer will have to charge, all else being equal.
I would also argue that there are circumstances in which residual interest really looks punitive. Consider the situation if I have 21 day grace and a $10,000 balance and pay $9,999.99 at the end of it and there are 9 more days in the cycle, during which time I make no further charges on the card. The finance charge is based on an average daily balance of is based on 21 days at $10K and 9 days on $.01--interest is
levied on a balance of $7000.03. At 10% for 1 month, no compounding, that means $58.33 in interest instead of <$.01 (rounded up to $1 or $1.50 by most cardholder agreements). (Let's hope I have the math right this time!)
The idea of not having any float isn't so crazy. If there was no float, consumers would use debit cards to transact and would be more careful about use of credit. That might not be such a bad idea.
I'm not ready to throw in the disclosure towel and concur with Tom Perkins. We've never really given disclosure a proper chance to work. Consider simple disclosures situations--like unit prices in grocery stores--those work pretty well. It allows an unsophisticated consumer to literally compare apples to apples. The question is whether we can make credit pricing fit into that mold. I don't think that simply amending TILA will do the trick. Instead, if we want disclosure to be effective, we will have to regulate price structures (not actual price) so that we can make it easy to compare total costs of credit between products.
I would also argue that group A should care about disclosures for two reasons. (1) there are a lot of folks who straddle the A & B group---occasional revolvers or, as I heard a card industry executive call them "sloppy payers," and people often underestimate the likelihood that they will be in group B (e.g., forgetting to put the check in the mail), and (2) transacting isn't free--there are substantial costs to merchants for each transaction, which are passed on to consumers. Even pure transactors should want to know these costs. (I recognize that interchange cannot currently be passed on to cardholders only in the US.)
Posted by: Adam Levitin | October 28, 2008 at 01:16 PM
Adam,
Perhaps we should require the following for people who have paid their last month's debt in full and might plan to pay part of their credit card debt this time:
"If you plan to pay less than the balance this month, you will be liable for interest of xxxx."
Posted by: Allan | October 28, 2008 at 03:42 PM
Last month I mistyped the amount when I paid my American Express bill. I paid $1 less than I should have to pay off the entire amount - $765.15 instead of $766.15. Given the way they compute the average balance they charged me $5.20 interest. I make it that they charged me 520% for a month to borrow the $1. That would be over 6000% annualized?
Posted by: KS | October 31, 2008 at 09:02 PM
"calculating what the cost of revolving a balance is just not realistic for even diligent consumers"
Which is why you shouldn't carry a balance. $0 finance charge is very simple.
If you are going to voluntarily carry $9,607.61 at 14.9%, it's silly to complain about the difference between the daily balance of $9,607.61 and $9,620.44, as that is a difference of 0.13%.
Yes the Banksters are swine, yes they intentionally obscure this stuff, but still!
Posted by: bobn | November 02, 2008 at 03:44 PM
bobn--obviously not carrying a balance is the way to avoid issues in the calculation of the finance charge. But even for those who don't intend to revolve a balance, there are lots of ways to slip up--forgetting to put the check in the mail, the post office taking a while to deliver, the issuer not processing the check immediately, etc. One day late (or even getting the check in on the due date, but not before the issuer's cut-off time) and there's a balance being revolved. There are an awful lot of "sloppy payers" around.
i think the importance of Lynn LoPucki's complaint isn't so much about the extra few dollars in interest---it's the cumulative effect. To illustrate, assume there are 500 million credit cards in force (there are actually far more--between 700 and 800 million). Also assume that half of those cards revolve balances at any point in a year (far more actually do), and the impact of residual interest over a year is $5 per card (again, a low estimate). That's $1.25 billion annually in extra interest due to residual interest. That's not chump change.
Posted by: Adam Levitin | November 02, 2008 at 05:16 PM
I got hammered by this little practice a few months ago and am not as livid about the interest and the way it's calculated as I am about the arbitrary "late payment fee" of $39 - and the "minimum finance charge" of $1 the following month on the few cents of interest calculated on the next statement on the "balance".
I made the mistake on a vacation of buying $49 of gas on a card that had a payment date close to my return date. I made all of my other charges on a card that I knew would not be due well until I returned.
I returned from my holiday a couple of days before the card's payment date and immediately put the $49 payment in the mail.
Well, the payment was credited a day after the payment date which triggered the residual interest calculation.
Needless to say I am NOT using this card any more. If next month's statement arrives with yet another $1 "minimum charge" on zero purchases then I will send the card cut into tiny pieces to the president of the company ( Sears ) which promotes this crap.
I have no debts and treat the cards as "convenience" cards. I am seriously considering "cash only" purchases after this experience of predatory lending at its finest.
Posted by: M Lipford | November 11, 2008 at 04:29 PM
Despite my best efforts to always payoff my credit cards in full I sometimes fail to do so. And so I fall into the residual interest trap. Before cancelling the card I call customer service and attempt to get the charges waived, which sometimes works. When it doesn't I make sure to overpay my next bill by 50 cents or a dollar. I then stop using the card. This generates monthly statements costing the credit card company 30 cents (?) to tell me that I have a 50 cents credit. It serves no useful economic purpose but it does assauge my residual desire for revenge.
Posted by: Walter Rosenkrantz | November 17, 2008 at 01:52 PM