Only Two Sources of Credit

The frame

"In almost every financial system there are only two sources of credit: it comes from the banking system or from the investment marketplace. There's no third choice. Around the world, for whatever reason, regulators have decided that banks should do less and investors should do more." Marc Rowan.¹

The move being made

This is a framing device before it is a fact, and the framing is the point. The public argument about private credit is usually posed as whether the industry should exist at its current size. Rowan reposes the question as, given that an economy needs a fixed quantity of credit, which of the two available containers should hold it. Once the question becomes venue rather than volume, several things follow: growth in non-bank credit reads as relocation rather than risk creation, shrinking the banks without expanding investors does not reduce credit risk but reduces credit itself, and the right regulatory question becomes which container suits which kind of liability rather than how large the second container is allowed to get. He restates the point later in the same interview at its most compact: "credit can come from investors or banks."¹

The Europe case is the test

The reason this framing carries real weight is that Rowan attaches a falsifiable comparative claim to it. Bank lending is under 30 percent of the credit market in the United States and still roughly 65 percent in Europe.¹ Pressure on banks has if anything been higher in Europe, through the Basel III endgame rules and similar measures, but the investor side has not been liberalized to match: "In Europe they are squeezing the banks harder than in the US, Basel III endgame and so on, but at the same time they have not liberalized on the investor side. It does not surprise me that Europe suffers from a capital deficit." The claim identifies a specific mechanism rather than stating a preference: the two containers are regulated by different authorities on different timetables, so a jurisdiction can tighten one without opening the other and end up with less total credit than either regime intended on its own. Nobody chose the resulting capital deficit; it is the residue of two uncoordinated decisions. It is also, plainly, an argument for letting Rowan's own industry operate more freely in Europe.

What makes an asset suit one container rather than the other

Rowan's allocation rule runs on duration and complexity rather than credit quality: "A bank is funded short and lent long. A bank is not a great source of really long-term capital. And think about what we're borrowing money for today: infrastructure, energy transition, next generation data and power, all really long-term assets. These are not the ideal assets for a bank balance sheet. Typically they would go to the investment grade bond market, but they happen not to be great assets for the investment grade bond market either, because they are generally complex projects or structured in some way." The modern financing need falls through both public containers at once: too long for a bank, too complex for a plain public bond. What a firm like Apollo does instead, in his words, is matching really long-term insurance liabilities with really long-term investment grade counterparties. What banks keep, stated without condescension, is "anything that is really operational or client centric."¹

The systemic arithmetic

Rowan draws a further conclusion from the same premise: "Every dollar that moves out of the banking system increases the resiliency of the system and reduces the leverage. A bank is levered 12 to 14 times. An investor is generally not levered at all."¹ He pairs this with a three-part negative definition of what the investment side does not do: no maturity transformation, no treasury guarantee, no deposits.

Tensions

The binary is doing more work than it can fully bear. Public bond markets function as a third container with different properties from either of the two named, and Rowan treats them as part of "the investment marketplace" when convenient and as a separate thing when convenient, shifting the taxonomy inside the same answer. The binary also hides the interfaces between the two containers, since banks lend to the funds that in turn lend to companies, and Rowan's own account elsewhere concedes banks remain in the structure at a very safe tranche, which means a two-container model with a pipe running between them is not quite the same claim as two genuinely separate containers. "Regulators have decided" is also passive in a way that understates how contingent the outcome was: no regulator wrote down a stated objective that banks should do less and investors should do more, and the sentence's framing makes an aggregate, uncoordinated consequence sound like an endorsed policy. And the claim that the United States is where half the world's capital gets raised is offered without a supporting figure, in an interview where the same speaker is elsewhere shown to vary his round numbers.