This piece is adapted from a research note by Kroll Bond Rating Agency.

The late Yogi Berra once famously said, “It’s déjà vu all over again.” The expression is apt for today’s financial sector.

The zero or low default rates being reported by banks may on the face be a good sign. But they in fact suggest mounting future credit risk, based on our most recent review of aggregate financial statement and portfolio data published by the FDIC. Credit does have a cost, so when U.S. financial institutions publish data that suggests otherwise we believe that investors and regulators need to beware.

During the height of the mortgage bubble in 2004 and 2005, banks such as Washington Mutual and Countrywide actually reported negative defaults. That means recoveries exceeded charge-offs, suggesting that credit had zero cost. During this same period, large financial institutions stopped preparing default studies because of the scarcity of credit events and the low levels of “loss given default,” or LGD.

Plainly stated, recent credit results measured by metrics such as charge-offs and LGD – defined as charge-offs less recoveries – are simply too good to be believed or sustained. Rising risk in bank loan portfolios results from the policies of the Federal Reserve Board, which explicitly set a strategy to push up asset prices to facilitate greater risk-taking. These higher asset values, however, have not been validated either in terms of rising income or gross domestic product. When valuations for commercial and residential assets are growing faster than the broader economy, and loan-to-value ratios are rising as well, the only certainty is that the prices will not be maintained.

In our last research comment on the banking sector, we noted that industry credit costs probably bottomed out in the third quarter of 2013 at just over $5 billion. The last time the banking industry’s provisions expense was that low was the first quarter of 2006, after which it rose sharply, reaching $71 billion by the fourth quarter of 2008. Loan charge-offs for the industry as a whole peaked a year later.

While banks don’t report credit loss provisions by loan category, they do report charge-offs, recoveries and delinquencies by loan type. Provisions were a leading indicator of future losses in the late eighties, the 2001 period, and run-up to the most recent crisis. Starting in 2007, banks aggressively raised provision levels above gross charge-offs to prepare for the worst period of bank credit losses in U.S. history.

In the period since the crisis, a combination of Fed monetary policy and outright asset purchases have again skewed the credit markets, shifting investor risk preferences in such a way that weaker issuers have been able to access the credit markets and create exposures that, in a “normal” interest rate environment, could be problematic. The sharp increase in leveraged lending volumes, especially in sectors such as oil and other commodities, has created the potential for significant future credit problems as asset values fluctuate.

The added external factor of falling commodity prices increases concern because of the implication for the financial solidity of banks and other financial institutions as well as the vendors and industries which serve them. At an investment conference this past summer in Jackson Hole, John Cushman, founder of the global commercial real estate firm that bears his name, talked about how commercial property valuations in cities such as Houston are “flying straight into the ground” due to low oil prices.

Depressed market prices for oil, copper and other key industrial commodities have cut the value of this collateral by more than half over the past year, raising the prospect of increased corporate defaults and related collateral damage for lenders and investors. As a result of factors such as Fed monetary policy and changes in commodity prices, we believe that a year from now default rates on many loan categories on bank balance sheets are likely to be higher.

Christopher Whalen is senior managing director and head of research at Kroll Bond Rating Agency