Let’s focus closely on the economics of the NICK business model. Here is how average transaction for NICK works, roughly:

The dealer buys a used car for $7,500 and sells it for $10,000. The dealer provides the buyer with financing for around 90% of the retail value of the car. The loan to the car buyer typically lasts for 4-5 years, and carries an APR (excluding ancillary fees) of 23% to 24%. The interest payments over the life of the loan total $5,000 to $6,000, meaning the car buyer owes the dealer about $15,500 in total. Spread over 4.5 years, this equates to a ~$300 monthly payment by the car buyer.

NICK is able to juice the return it earns on loans by buying them from the dealers for a slight discount, typically for an 8-9% discount to principal. The exact discount fluctuates based on the intensity of industry competition at the time. Right now competition for loans is fierce and NICK is buying them for a 7.5% discount; in the wake of the GFC, when competition was dramatically less, NICK was getting discounts in excess of 9%.

The discount means that, on average, NICK pays the dealer about $9,100 for $10,000 of loan principal.

To recap, the dealer paid $7,500 for the car, and after the sale of the car, the dealer received $9,100 cash from NICK. That means the dealer’s gross profit is about $1,600, give or take a couple hundred dollars depending on the specifics of the deal.

NICK has paid $9,100 for a $10,000 loan paying 23-24% interest. Because of the purchase discount, as well as add-on fees, the effective gross yield NICK earns on its $9,100 ends up being 27-29%, rather than 23-24%.

The numbers I’ve presented above are indicative of an average transaction for NICK. For other lenders the numbers can be different. Loans for new cars and/or prime borrowers are typically larger in size, longer in term, and lower in interest rate. Not all lenders buy the loans from the dealer at a discount either.

*          *          *

Now let’s take a brief diversion and talk about NICK’s accounting. I’m going to discuss the accounting in terms of the portfolio as a whole, but the concepts apply to individual loans as well.

You can see the individual components of NICK’s loan portfolio in the following table, which is contained in every 10-K and 10-Q that the company files:

(The table above shows loans NICK has purchased from dealers. NICK also has a very small portfolio loans it has originated directly to borrowers.)

We’ll begin with the first item, “Finance receivables, gross contract”. This represents the entire principal amount loaned to the borrower, plus all of the contractual interest to be paid over the life of the loan. If NICK were never to make another loan, and never to lose a single dollar on any of the loans it had already made, this is the total amount it would receive over the remaining lives of all the loans in its portfolio.

The next line, “unearned interest”, is exactly what it sounds like. It represents all the contractual interest that remains to be paid over the lives of the loans. This interest is recognized over the life of the loans as revenue.

If you subtract “unearned interest” from “finance receivables, gross contract”, you get “finance receivables, net of unearned interest.” This is the gross amount of original loan principal in the portfolio. This is the metric NICK uses in its portfolio summary (discussed a few paragraphs below) to measure the ongoing performance of its business.

The next line item is “unearned dealer discounts”. When NICK buys a $10,000 loan from a dealer for $9,000, the $1,000 difference goes into this account, and then gets amortized onto the income statement as revenue over the life of the loan.

Then, we have the “allowance for credit losses”. This balance sheet item represents NICK’s estimate of the losses it will incur on its current portfolio over the next twelve months. Changes to this line item run through the income statement as provisions for losses.

I want to note here that I think there are two ways to evaluate NICK’s loss reserves. The first is to simply use this line item, the allowance for credit losses. But another rational way is to include the dealer discount on top of the credit allowance, since technically NICK’s loan losses do not eat into net principal until both the credit allowance and the dealer discount have been exhausted.

Finally, we have “finance receivables, net.” That means net of everything—unearned interest, unearned dealer discounts, and loss allowances. “Finance receivables, net” is what NICK reports on its balance sheet.

That is a quick summary of the accounting, which will help when we review NICK’s economic model.

*          *          *

Now, how does NICK’s business work over an economic cycle?

I think the best way to evaluate NICK’s business performance is to measure income statement items against the loan portfolio, i.e. “finance receivables, net of unearned interest.” NICK provides this analysis in its filings, in a “Portfolio Summary” table that looks like this:

In this analysis, there are four key metrics:

1. Gross portfolio yield. This equals revenue divided by the average receivables portfolio over the period. This yield is greater than the “weighted average contractual rate” (aka the APR, or interest rate on the loan) because it incorporates the discount at which NICK buys the loan from the dealer, as well as add-on services and fees paid for by the borrower.

A high gross portfolio yield indicates NICK is making loans at the maximum possible rates, likely because there is a paucity of competition. A declining gross portfolio yield—like we are seeing today—indicates increased competition among lenders, and thus better terms for NICK’s counter-parties (i.e. lower rates for borrowers and lower discounts from dealers).

2. Provision for loan losses. This represents the income statement expense the company takes in the period in order to maintain what it believes is a fair estimate of future period uncollectible loan principal and interest. (The loan loss provision rate is different from a charge-off, which represents the receivables NICK actually wrote off for good as uncollectible during the period, and which is a debit to the loss allowance on the balance sheet).

An increasing loss rate can be driven by two separate factors. The first factor is a deterioration in general economic conditions, such as occurred in 2008 and 2009. Job losses increase, wages decrease, and existing borrowers become less able to pay off their loans.

The second factor is the quality of newly originated loans. If new loans are being made on less favorable (to the lender) terms, or to less creditworthy borrowers, then the portfolio as a whole will experience increasing losses as the older, higher-quality loans mature and are replaced by new, lower-quality originations.

I believe NICK’s (and the industry’s) recent increases in loan losses are exclusively due to this second factor. The US economy has been improving since the financial crisis. This economic improvement, while probably improving underlying loan quality on its own, has also increased the availability of capital to the point where competition among lenders has reached extremely high levels. In response, lenders are loosening income requirements, allowing longer payback schedules, and lending below state maximum APRs, and these actions are driving increased loan losses.

3. Interest expense. This represents NICK’s cost of debt capital, multiplied by the amount of leverage used to finance the loan portfolio. For instance, if NICK’s cost to borrow funds is 4.5%, and NICK finances its book of receivables with two dollars of debt for every dollar of equity (67%), then the interest expense ratio is 3% (4.5% x 67%).

4. Overhead. This should be self-explanatory. In NICK’s particular case, it is worth noting that this line item has recently included considerable expenses from the aborted sale of the company and from Canadian taxes related to a large special dividend the company paid. Neither of these costs should recur, although the company will incur some level of ongoing professional fees, perhaps equal to 30 basis points of the loan portfolio annually.

Below, you can see how NICK’s expenses, individually and together, have trended over time as a percentage of the loan portfolio:

Overhead has been stable at around 10% of loans receivable, although it seems to be moving up slowly over time despite NICK’s increasing scale, presumably due to increasing costs from compliance and regulation. In addition, NICK’s branch-based, high-touch model is less scale-able than its competitors’ quantity-based business models.

Interest expense has decreased over time. This is a function of both decreasing debt costs worldwide, and of NICK’s lengthening track record as a reliable borrower (and thus increased creditworthiness in the eyes of the banks who provide its credit line). Note that NICK’s interest expense will increase about 35% (70 basis points) going forward because of the new, higher leverage ratio as a result of its debt-funded tender for 40% of its shares.

Finally we have loss provisions, which as you can see are quite volatile, and quite high at the moment due to the competitive environment.

Now let’s add revenue/profit. Here is the same chart, but now with what’s left over—pre-tax profit—at the bottom:

You can see that revenue (i.e. the gross portfolio yield) is typically 25-30% of receivables. The profit yield is cyclical, but putting aside 2009, has remained fairly steady across the cycle.

Throwing out the best (2012) and worst (2009) years, pre-tax margin has remained within a band of 7% to 12%:

Over the cycle (which I measure as six years here because most of the recent cycles have lasted roughly that long), the average pre-tax margin has been approximately 10%.

In the next post I will discuss present business conditions in further detail.