WASHINGTON — Policymakers pushing to scale back regulation have relied heavily on a core argument — bank lending is being held back by post-crisis capital rules and other restrictions.

But assessing whether that is actually true is more difficult than it appears.

“It’s not the kind of issue that lends itself to crisp, absolute answers, unfortunately,” said Greg Lyons, a partner at Debevoise & Plimpton. “It’s a more amorphous and an ideological view as much as anything else. There’s not necessarily a right or wrong answer. It depends on what your objectives are.”

Lending data 2016 for small businesses

Federal Reserve Board Chair Janet Yellen has repeatedly said that banks' lending activities have not been appreciably affected by new rules. In a hearing last month, she pointed to a survey of members by the National Federation of Independent Business which suggested that only a small fraction of small businesses are unable to get the credit they desire.

But loan demand is a difficult thing to measure, in part because it’s hard to count loans that aren’t made. By some metrics, lending demand is down.

The Fed's H.8 data, which details the size and relative growth or contraction of commercial banks’ assets and liabilities on an aggregate basis, showed that banks’ overall credit availability grew at around 7% per quarter between 2014 and the third quarter to 2016, dropping off to a little under 5% on average after that.

The most recent senior officer loan survey, meanwhile, reported that bankers thought demand for commercial and industrial loans and commercial real estate loans had dwindled, running counter to their expectations earlier this year. The survey also found that bankers felt that lending standards are generally looser now than they were in 2005, theoretically making it easier for borrowers to obtain credit.

Nellie Liang, the Miriam K. Carliner Scholar in Economic Studies at the Brookings Institution and former head of the Fed’s Office of Financial Stability Policy and Research, said in a speech in July that while higher capital levels can reduce banks’ returns on equity, the effect in the U.S. is not as dramatic as some would have it appear. In part, that's because comparing pre-crisis to post-crisis credit availability leaves out one important factor: the costs of the financial crisis.

“To be sure, credit is probably more expensive compared to risk-free rates—a healthy development considering how overstretched the pre-crisis financial system was—but the effects on flows appear to have been small,” Liang said. “While it is not clear how much credit is necessary to support economic growth, it is clear, however, that it was too high before the crisis, so growth rates at that time are not the right benchmark.”

But the industry says there is a worrisome trend that those aggregate numbers fail to address, and that is the heightened challenges that banks — particularly larger institutions — face in lending money to borrowers with less than perfect credit scores.

Bill Nelson, chief economist and head of research at The Clearing House Association, said that the new capital and liquidity rules — and especially the Fed’s stress testing program — have made it especially hard for the largest banks to make loans that run a risk of default under economic stress.

“The stress tests achieve their stress by assuming a very severe macroeconomic downturn,” Nelson said. “By [definition] that is going to lead banks to substitute away from loans that are exposed to such downturns.”

That has a particularly negative effect on the formation of small businesses, said Francisco Covas, senior vice president and deputy head of research at The Clearing House. While the dollar amounts to get small businesses off the ground are relatively small, they account for a disproportionate segment of job creation and economic growth, he said.

“That really impacts the long-term dynamics of economic growth," Covas said. "These are regulatory changes that, even though the numbers on lending are not huge, they are potentially very important because they are important for economic growth.”

Covas said small businesses are often started with home equity lines of credit or other noncommercial vehicles — areas of lending that the H.8 data shows have been contracting for years. Commercial lending rates may seem stable and healthy, he said, but that may only reflect the market for loans to well-qualified borrowers who don’t need credit as badly.

“This is consistent with the narrative that, if you’re a very safe borrower or a large corporation, it is very easy for you to get a loan — in fact there is a lot of competition because these borrowers have access to lots of nonbank sources," Covas said. "However, if you are perceived to be a riskier borrower … it’s very hard for you to get credit.”

Karen Shaw Petrou, managing partner at Federal Financial Analytics, said there is reason to be suspicious of the top-line lending figures, because they tend to treat lending as equally beneficial even though not all loans have the same power to spur growth and counteract financial inequality.

“If you break down the data by how it compares to GDP, how it compares per capita, how it compares to key lending for economic inequality and recovery — not IPOs and underwriting — lending to corporations for investment, which is way down … a lot of lending is designed to fund dividend and share repurchases,” Petrou said.

Federal Deposit Insurance Corp. Vice Chairman Thomas Hoenig echoed that sentiment in a July 31 letter to Senate Banking Committee leaders, arguing that capital is what banks use to make loans, and rather than being too high, capital requirements may actually still be too low.

On an aggregate basis, banks are already passing on almost all of their quarterly earnings, or even more in some cases, to their shareholders in the form of dividends and share repurchases, Hoenig said. If those banks were to retain more of their earnings, they could use those funds to make the kinds of loans that banks say regulations are keeping them from making, he said.

“Despite rhetoric to the contrary, capital is not ‘idle’ and it does not ‘inhibit’ lending,” Hoenig said. “I recognize that dividends are an important factor for investors and they should be rewarded for the risks they take. But it is also true that funding business growth, assuring future economic success and promoting capitalism depends upon the retention of earnings.”

Lyons at Debevoise & Plimpton said the criticism of banks’ dividend payments and stock repurchases makes superficial sense, but banks are funded through shareholder investment. If one prefers to reduce a bank’s volatility through higher capital, that is a rational choice to make, he said, but it comes at the expense of the economic activity that banks can facilitate.

“In my mind, it’s a philosophical view. Do you want to let the banks do more lending to promote the economy, understanding that it could result in the banks being potentially more volatile, or do you want to ... have them have such high capital ratios that they can’t promote the economy to the same degree, but they are arguably more stable?” Lyons said.

Petrou said that a preference for stability is affecting the mortgage market as well, depriving poor people of an opportunity to own their own homes — the primary mode for poor people to enjoy the economic stability of the middle class. A paper her firm released on July 31 noted that while the mortgage market seems to have stabilized for wealthier Americans, the availability of credit for poorer borrowers has not improved, she said.

“If you segment housing markets by income or price, houses most people own have at best flattened, and the values of lower-income people have fallen by as much as 46%,” Petrou said. “And the credit availability for low, moderate and younger Americans is terrible.”

Nelson acknowledged that the solution to these problems is not to roll back capital rules or supervisory exercises entirely. Indeed, in the case of the stress test, he said, the concept of a “stress capital buffer” described by former Fed Gov. Daniel Tarullo and in place in other jurisdictions would ease some of banks’ biggest complaints about the stress testing program. But there is no sense in denying that the rules have had an effect on the markets.

“We think these are unintended consequences — we don’t think the Fed set out to try to get banks to make fewer small-business loans or to not make loans to households that have very good credit scores but not perfect ones,” Nelson said. “We think these are just flaws in the design of the regulations.”