Oil flow causes stress

By Zach Fox, SNL Financial staff writer

As banks were finalizing stress test submissions for an April 5 deadline last week, a quick glance at last year's exercise shows banks could face a very different 2017 and 2018 than what the Federal Reserve outlines in its various scenarios.

In the last two stress tests, the Fed's base-case scenarios have overshot interest rates. And in last year's stress test, the Fed's severely adverse scenario asked banks to consider what loan losses would look like if oil reached $110 per barrel. Since then, oil has collapsed, bouncing around $30 per barrel to start the year and presenting an entirely different set of issues for banks and the economy.

While stress tests are not intended to be economic forecasts, the disparity between the Fed's scenario and reality underscores the difficulty of ensuring banks are prepared for the next crisis.

"One thing that people point out time and time again is that if you only looked in the rear-view mirror in 2006, you would never expect a 30% decline in home prices," said David Wright, a managing director for Deloitte & Touche LLP.

The Federal Reserve declined to comment.

Concepts behind stress tests

The base-case assumptions, by design, represent conventional wisdom as an average of various economic predictions from forecasters. The adverse and severely adverse scenarios then stress bank holdings with a variety of economic shocks.

Despite the differences on some variables, Wright said the stress tests do sufficiently evaluate the stability of bank balance sheets under adverse economic conditions. While banks are bracing for notable losses due to the collapse in oil prices, a spike in oil prices could be worse. Higher gas prices can strain consumers, driving up losses in consumer and mortgage loan books that are generally much larger than oil and gas holdings.

And oil prices represented only one of many variables. The severely adverse scenario included a spike in the unemployment rate to 10.0% by the 2016 first quarter, whereas the figure clocked in at 4.9% in February.

"The hope is that all the other shocks that you have in the stress test cover that [oil price] shock," said Chuck Morris, a vice-president and economist for the Federal Reserve Bank of Kansas City. "We had a big decline in oil prices, but there are other stress-test shocks that didn't happen."

As further protection, the Fed makes each bank devise an idiosyncratic scenario that stresses its unique vulnerabilities. Presumably, banks with heavy oil and gas exposure would devise scenarios that include a collapse in energy prices, Wright said. Also, the Fed monitors oil and gas losses through regular reviews of the nation's largest loans through the shared national credit program.

"Of course, [the shared national credit program] has been focused on oil and gas for quite some time now," Wright said.

Negative rates hit

This year, the Federal Reserve threw banks a curveball in the stress test by including negative interest rates, a headwind banks in Japan and Switzerland currently face.

Also, the Fed's scenarios are more severe on a relative basis. For example, the unemployment rate spikes by 4 percentage points in the 2016 exercise, compared to a 2.1 percentage point jump in the 2015 exercise. However, because the economy is stronger, the increase is applied to a lower base, meaning the peak unemployment rate of 10.0% in this year's stress test is lower than any of the previous three tests.

Even with a more rapid deterioration in the economy and the specter of negative interest rates, analysts said they expect banks' balance sheets to withstand the Fed's most severe scenario.

"The fact that capital has been built up year after year after year has made it unlikely that firms will fail on a quantitative basis," said Michael Alix, a partner with PwC.

However, Alix and Deloitte's Wright both said it was possible banks could fail the qualitative portion of the Fed's exercise, which evaluates a bank's risk management procedures. That was the case last year, when the U.S. subsidiaries of Banco Santander SA and Deutsche Bank AG were able to pass the quantitative stress test but failed on qualitative issues such as governance and internal control deficiencies.

Getting the numbers accurate

This year, the Fed has made it very clear to banks that there will be a focus on data quality.

"'Where did you get your data? How good is it and how reliable is it?' My clients are getting a lot more questions about their data practices," said Mayra Rodriguez Valladares, a managing principal of bank consulting firm MRV Associates.

Alix and Wright both agreed, saying data quality appears to be an area of focus for the Federal Reserve in this year's stress tests.

"Firms that don't have a buttoned up process for reconciling and reporting the data … will be more vulnerable to a fatal problem," Alix said.

It is hard to imagine the quantitative portion of the exercise could prepare banks for a repeat of the 2008 credit crisis and an unexpected collapse in a certain asset, such as housing. And it is an open question whether bracing for highly unlikely events would be wise since it would almost assuredly constrain access to capital.

Considering the difficulty—perhaps impossibility—of ensuring bank balance sheets can weather the next crisis, the stress test's value might be in the qualitative assessment.

The views and opinions expressed in this piece are those of the author and do not necessarily represent the views of S&P Global Market Intelligence. Share your thoughts at [email protected].

This article originally appeared on SNL Financial’s website under the title, "Last year's Fed stress test included $110 oil, showing limits of exercise."

Download reprint of SNL article