Since the financial crisis, the once-intense battle between full-service mega-banks like Citigroup and JPMorgan and traditional investment banks like Goldman Sachs and Morgan Stanley has become less relevant. When every bank is struggling to survive, two groups competing for dominance seems less relevant. But now that the industry has mostly healed, the war wages on. This week, we learn of a new twist that could dramatically help the investment banks: a new rule is being considered that would threaten corporate "relationship banking."

First, a little explanation here is needed for readers who aren't familiar with the term. When the Glass-Steagall Act passed after the Great Depression, institutions could either take deposits or perform securities work like issuing stocks and bonds, but not both. That separated the world into two types of banks. One type specialized in consumers and/or commercial customers. The other specialized in securities transactions and market making.

Then, in 1999, the Gramm-Leach-Bliley Act passed and repealed that rule. This excited depository institutions, which began acquiring investment banks. For example, Citicorp acquired Salomon Brothers and Chase Manhattan acquired J.P. Morgan. Today, we have Citigroup and JPMorgan Chase. Not all investment banks got acquired, however. At the time, five big players remained independent, which included Goldman Sachs, Morgan Stanley, Merrill Lynch, Lehman Brothers, and Bear Stearns. Of course, the financial crisis eliminated those last three, with JPMorgan acquiring Bear, Bank of America acquiring Merrill, and Lehman failing. A number of small boutique investment banks also still exist, however.

"Relationship banking" became one of the big advantages of the full-service institutions. This is a sort of an euphemism to describe an ongoing client relationship where a bank promises a firm below-market rates on lending if that firm provides the bank its securities business. For example, perhaps a restaurant chain wants to expand nationally, but it needs a number of commercial mortgages. Citigroup could step in and provide it an interest rate 1% below the prevailing market rates if the firm promises to allow Citigroup to be the lead manager on its stock offering. Full-service banks refer to their superior "balance sheet" compared to investment banks, which means that they have more capacity to lend.

As you can imagine, this annoys the investment banks to no end. They can't offer the same attractive deals to clients because they don't have a balance sheet a big as a major depository institution would. Instead, they must rely on their expertise. They've managed to do pretty well over the years on their reputation alone, but they've lost many billions of dollars in business to the full-service banks as well due to relationship banking.

And finally to the news: the Financial Times reports that the Financial Accounting Standards Board is considering a new rule that would make relationship banking more costly for the full-service banks. Helen Thomas at FT writes:

The proposed rules, on what is known as initial measurement, would require banks to make clear where the transaction price of a loan differs from its fair-market value. In some cases, the difference could hit a bank's earnings.

These formed part of the US standard-setter's proposals on fair-value accounting, released last year. The body in January backtracked on a broader move to require banks to value their loan books according to market prices.

What does this mean in English? Let's say that, in the example above, Citi provides $500 million in commercial mortgages at 1% below the prevailing market rate. This implies that the loans will have more risk than usual, because the prevailing market rate should take into account the proper cost of risk. As a result, Citi might be forced to declare a loss on these loans under the new rule. If it sold those loans, an investor would demand a discount due to the below-market interest rate.

This idea sort of makes sense, but it's hard to see how it could actually work in practice. In the example above, we're talking about a company that is not yet public at the time the loans are made. It almost certainly isn't rated by the rating agencies. How do you establish what the right risk-based market pricing would be for its loans? You could try to back into it using comparable firms' loans, but often such terms won't be public to consider. Such loans also often won't be commonly traded, as the bank can choose to retain them on its balance sheet, since there might not be a secondary market for the loans in all cases. So how do banks or the accounting regulators know where the market would stand on these illiquid loans?

This relates to the more generalized problem of broad mark-to-market requirements, when loans are forced to be valued on a bank's balance sheet at their market value. But in this instance, it might be even more difficult because so many of these loans are likely to be highly customized and rarely traded. So while this new rule might seem fair, it's pretty hard to see how it could actually be used. We'll have to wait to see how accounting regulators explain its implementation, if they approve it.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.