Today banks are giving relationship managers and partners more room to set exception pricing for customers. A U.S. car-lending case shows what that room can cost when nothing limits it. Between April 2011 and March 2012, African American borrowers at Ally Financial paid on average 29 basis points more in dealer markups1 than comparable white borrowers. The disparity was not necessarily apparent from an individual loan; it emerged when regulators analyzed more than 800,000 loans collectively. The reason could be traced back partly to the discretion given to individual car dealers over pricing. Dealers could add up to 2.5 percentage points to a loan rate the lender had already approved and receive compensation from the additional interest. The higher the markup, the more the dealer earned. The markup was determined at the signing table, giving the lender limited visibility into individual pricing decisions.
In December 2013, the Consumer Financial Protection Bureau (CFPB), the U.S. regulator for consumer finance, and the U.S. Department of Justice ordered Ally to pay USD 80 million2 to approximately 235,000 borrowers. About 100,000 African American borrowers had each paid, on average, more than USD 300 in additional interest.
What the Order Asked For
The order3 let dealers keep setting the markup, on three conditions.
- A limit: the markup stayed capped at 250 basis points, and lower on longer loans and weaker credit, so no single pricing decision could do unlimited harm.
- A regular check: every quarter, Ally had to analyze each dealer’s pricing for gaps between groups of borrowers, estimating race from surnames and addresses because loan files do not record it.
- A consequence: dealers whose pricing kept showing a gap could lose the right to set prices, and borrowers were to be refunded if the gap persisted.
Together, these conditions created a framework for controlling pricing discretion. Every departure from the lender’s approved rate could be identified and monitored. There were defined limits on how far the price could move, pricing decisions were reviewed regularly for problematic patterns, and there were consequences when those controls failed.
In a commercial environment, a salesperson may exercise similar discretion by offering a discount or concession. While the commercial circumstances and risks are different, the underlying governance principle is the same. When individuals are allowed to move away from an approved price, the organization needs to know the following:
- When an exception was made
- The extent of the exception
- If it is within established parameters
- What happens when those parameters are breached?
This effectively creates an exception register or a structured record of pricing decisions that fall outside the standard rate. In Ally’s case, the framework was imposed by a regulator and managed through compliance. Commercial organizations need to establish and enforce similar disciplines themselves.
Why It Matters Who Writes the Rules
In B2B sales, the team that approves prices outside policy is often called a deal desk. The firm writes the deal desk’s rules itself. These include key points such as how big a discount a salesperson may offer, and who must approve of anything bigger. Unlike externally imposed regulatory controls, however, these are internal rules, which means commercial pressure can sometimes lead to exceptions becoming more frequent or the boundaries themselves being relaxed. In case of Ally, dealers had no such room, because a regulator’s order cannot be renegotiated to close a sale.
Knowing Whether a Price Covers Its Cost
Ally’s exception register provides visibility into when a price deviates from established rules and whether that exception remains within permitted limits. But governance alone cannot answer if the resulting price still covers the cost of providing the product or service. This requires reliable cost information. For example, Ofcom, the UK’s communications regulator, requires BT, the country’s dominant fixed-line telecoms provider, to publish audited accounts4 that allocate costs across regulated markets and network components. This level of cost visibility makes it possible to assess whether the price charged for a service adequately reflects the cost of providing it.
What Can Be Copied and What Cannot
A firm could take one of two approaches:
- Use the exception records that its compliance team already maintains
Or
- Borrow the four questions that underpin those records:
- What constitutes a pricing exception?
- How far can the price deviate?
- Who reviews the exception and how often?
- What happens when established limits are breached?
The second approach allows commercial teams to apply the same governance principles without accessing or repurposing compliance records. There appears to be no publicly documented example of a firm taking either approach, so the following considerations represent a proposed framework rather than established precedent:
- Margin: The exception register can show whether a pricing decision is compliant with established rules, but it cannot determine whether the resulting price makes commercial sense. A discount can be fair, consistent, and fully compliant with policy while still giving away more value than the economics of the deal justify. Assessing this requires a reference price against which the discount can be measured and reliable information about the cost of providing the product or service. For some bank fees, one or both of these may not be readily available.
- Law: Regulation B5 is a U.S. regulation that prohibits discrimination in credit transactions on grounds including race, sex, and age. It also allows lenders to conduct self-testing to identify potential discrimination in their own pricing, with certain protections applying to the results of those tests. But these protections apply to new information generated by the self-test. They do not necessarily extend to underlying loan records the lender already maintains. An existing exception register would fall into the latter category.This creates a potential concern if a pricing team combines information from the exception register with profitability data. The resulting analysis may not receive the same protection as information generated through a protected self-test. And it could potentially be sought in litigation and used to examine whether different groups of borrowers were treated differently.There appears to be no law or court decision stating that creating such an analysis itself is illegal. Nevertheless, compliance teams responsible for managing regulatory and litigation risk may be reluctant to allow existing compliance records to be repurposed in this way.
- Scope:Regulation B applies to credit6, and involves transactions in which payment of a debt is deferred. A waived account fee, for example, is not itself credit, while U.S. rules governing fee disclosure7 and unfair fees 8 do not require banks to maintain an equivalent record of every fee discount.
Yet discounts across retail, corporate, and wealth management can exceed 50 per cent of headline prices in some cases9. When such discounts apply to fees or other prices outside the scope of these regulatory requirements, there may be no equivalent compliance record unless the bank creates one itself.Legal and scope considerations make a strong case for leaving the existing exception register with its compliance owners. But they do not prevent firms from applying the same four governance questions to pricing decisions that fall outside regulatory oversight. Firms must also focus on margin to measure the financial impact of each pricing exception alongside whether it complies with established rules.
Where to Start
List the prices where no outside party requires a record of every exception. For a bank, this includes fees and relationship pricing outside lending. Answer the four questions for each, with the understanding that rules a firm writes for itself can still be reopened.
Firms should also measure the margin given up with every pricing exception against a clearly defined reference price. If that margin cannot be calculated, the first priority should be to establish reliable reference prices and cost information. A pricing exception still has an economic impact even when that impact is not being measured. Either the firm absorbs the cost, or it may ultimately be recovered through higher prices elsewhere.
The next step is to establish clear limits on how far prices can deviate before determining the approval process for exceptions. The CFPB found that limiting dealer markups to 100 basis points, rather than the more common 200 or 250, “may reduce or even effectively eliminate” pricing disparities10 and potentially reduce the level of monitoring required. The importance of meaningful limits is also evident in commercial discounting with private banks approving more than 97 percent11 of discount requests. When almost every exception is approved, the approval process runs the risk of becoming just a mechanism for recording discounts rather than an effective control over them.
The BT and Ally examples demonstrate two complementary requirements for effective pricing governance. UK telecoms regulation required BT to maintain detailed cost records but did not create an equivalent pricing exception register. The U.S. lending regulation required Ally to monitor pricing exceptions but did not provide the cost information needed to assess their commercial impact. Effective discount governance requires both visibility and control over exceptions, as well as reliable information about what those exceptions cost the business.
That starts with recognizing that pricing decisions are being made far beyond the pricing function. Somewhere in the organization, people are deciding what gets recorded, how it is captured, and how often it is aggregated. Those may look like operational or data decisions. But they shape which pricing choices are possible later. They are pricing decisions too.
