The Federal Reserve's proposal last year to tailor regulation on large banking organizations based on their risk profiles included a plan to eliminate or ease the liquidity coverage ratio (LCR) requirement on 15 banks, representing almost 15% of assets in the U.S. banking system.

S&P Global Ratings believes the proposal, if enacted, could moderately reverse some of the improvement in liquidity that the U.S. banking system has shown since the financial crisis and conceivably could lead to a material negative change in the liquidity of the subset of banks no longer subject to the LCR. Banks subject to the LCR generally have had significantly stronger liquidity than other banks, and banks subject to the most stringent LCR calculation have typically had stronger liquidity than those subject to the less-stringent "modified" version, in our view. In other words, it appears the LCR has had a real impact.

On top of that, the elimination of LCR for a number of banks would weaken transparency for investors. The ratio provides insight not only to liquidity management, but also to the makeup of banks' funding. Although it's not our base-case expectation, it's possible we would lower ratings on banks that end up with substantially less conservative liquidity management after the elimination of the LCR, resulting in materially lower liquidity.

Background: What Is The LCR And How Might It Change?

The LCR aims to measure a bank's ability to meet a cash outflow assumed to occur during a 30-day stress period based upon the nature of its funding and other cash obligations. High-quality liquid assets (HQLA)--essentially unrestricted cash and very liquid securities--make up the numerator of LCR. Net cash outflows--the cash outflows netted for cash inflows assumed to occur during the stress period--make up the denominator. For example, a bank with a 100% LCR would theoretically meet all net cash outflows during the 30-day stress period and wind up depleting all of its HQLA.

U.S. regulators finalized the LCR rule in 2014, requiring banks under enhanced supervision (at the time, basically those with at least $50 billion of assets) to fully comply by Jan. 1, 2017. They required banks with at least $250 billion in assets or $10 billion or more in on-balance-sheet foreign exposure (and their subsidiaries with at least $10 billion in assets) to maintain LCRs of 100% by that time. Regulators also crafted a "modified" LCR for banks with $50 billion or more in assets, but less than $250 billion, simply by multiplying the net cash outflow (the denominator in the equation) by 70%. Those banks also needed a 100% LCR but were able to achieve it with less HQLA because of the less-stringent 70% assumption embedded in the denominator. As smaller institutions, these banks pose less systemic risk and therefore have the lower requirement.

The story changed somewhat in 2018. First, in May 2018, President Donald Trump signed regulatory reform legislation (The Economic Growth, Regulatory Relief, and Consumer Protection Act), raising the threshold for banks subject to enhanced supervision. That bill set the new threshold at $250 billion for U.S. bank holding companies but gave regulators the discretion on how to treat banks with at least $100 billion in assets. In effect, that initially lifted the LCR requirement, as well as a number of other regulatory requirements, for the few banks with assets between $50 billion and $100 billion.

The Fed followed up in October 2018 with its plan to tailor regulation on large banking organizations based on their risk profiles (see "U.S. Regulators' Plan To Ease Bank Rules Is Largely In Line With Our Expectations," published Nov. 7, 2018). It basically proposed to eliminate the LCR generally for domestic banks with assets between $100 billion and $250 billion and to reduce the LCR requirement to 70%-85% (rather than 100%) for non-globally systemically important banks (GSIB) with assets of $250 billion to $700 billion. (The Fed also considers other risk factors, but asset size is the most prominent driver for determining the LCR requirement).

Positively, the Fed would still require these banks to carry out internal liquidity stress tests and maintain certain liquidity risk management standards. It is possible that the internal stress tests could be more stringent than the LCR in some cases, but that is unclear at this point.

The LCR Has Helped Support Liquidity In The U.S. Banking System

Following the financial crisis, liquidity in the U.S. banking system improved sharply from precrisis levels. The Fed's quantitative easing (QE) program--occurring over three rounds between 2008 and 2014--probably played the greatest role in boosting liquidity in the banking system. The Fed increased its assets to about $4 trillion from less than $1 trillion during that period, contributing to a roughly equivalent increase in cash and securities held by banks. Since the end of QE in 2014, liquid assets have grown at a much more modest pace and actually have fallen since hitting a peak in 2017--probably partially resulting from the Fed shrinking its balance sheet.

Chart 1

While QE Boosted Liquidity, The LCR Has Helped Force Banks To Maintain It

The existence of the LCR requirements has had a significant effect on the amount of liquidity individual banks hold, in our view. Federal Deposit Insurance Corp. (FDIC)-insured commercial banks with assets greater than $100 billion had the greatest increase in liquidity as the Fed carried out QE (see chart 1). They also have maintained that improved liquidity. Banks with less than $100 billion in assets also had a notable increase in liquidity from 2008 to 2012 but have since seen a decline by some measures.

The 2013 proposal and 2014 finalization of the LCR rule likely helped cause the large banks to continue building liquidity beyond 2012, while other banks perhaps were able to curtail liquidity to some degree. What's more, those banks subject to the most stringent LCR requirement have generally built more liquidity than those subject only to the modified version. To be fair, the LCR probably has not been the only driver of the differences in liquidity between large and small banks. It is possible that the Fed's QE had a larger impact on corporate deposits--which are probably held disproportionally by the large banks--than on retail deposits, and thereby increased liquidity held by the large banks more than the small ones. QE in effect added to deposits to the banking system as the Fed bought securities. The proceeds of that spending ended up as deposits on the liability side of bank balance sheets (and effectively liquid assets on the asset side).

Chart 2

Banks Not Subject To The LCR Generally Have Lower Liquidity Than Those Subject To It

Because of insufficient disclosure, it is difficult to estimate what the LCRs would be for banks not subject to the requirement. However, we can calculate other liquidity ratios, which indicate that banks not subject to LCR generally have weaker liquidity than LCR-compliant banks.

Some of the banks we rate have significantly weaker liquidity than peers, which factors holistically into our ratings (see chart 3). However, some of these banks have offsetting factors that these ratios do not capture--such as significant holdings of conforming residential real estate mortgages that could easily and quickly be sold into Fannie Mae or Freddie Mac mortgage-backed securities.

Chart 3

The LCR Provides Valuable Insight Into Banks' Funding

Banks subject to the most stringent calculation of the LCR have publicly disclosed their calculations of the ratios since second-quarter 2017. Those reporting modified LCRs only began disclosing since year-end 2018. Those disclosures provide insight into which banks have the greatest and least sources of liquidity (mainly HQLA), sources of liquidity risk, ratios of sources of liquidity versus liquidity risk, and best deposit quality.

For instance, Northern Trust reports both the greatest sources of liquidity and liquidity risk, in relation to its assets (see chart 4). The other two trust banks, Bank of New York and State Street, also have relatively large sources of liquidity and liquidity risk--the latter relating in large part of high levels of wholesale uninsured deposits. The banks focused on credit cards (American Express, Discover, and Synchrony) show some of the lowest sources of liquidity and liquidity risk. That is in part because the LCR incorporates an assumption of no liquidity outflows on unused consumer revolving lines with the premise that they can be canceled at the company's discretion.

Chart 4

The LCR disclosures also provide detail on what drives each bank's sources of liquidity risk. Wholesale deposits--particularly nonoperational wholesale deposits--and credit and liquidity facility commitments represent the greatest liquidity risks for most banks subject to the LCR. For banks with significant broker-dealer operations, an outflow relating to secured wholesale funding is a major driver of the LCR. However, these banks generally also have significant estimated inflows from secured lending. (The rate of inflow and outflow depends on the asset securing those borrowings and loans. For instance, there is a 0% outflow assumption on borrowings secured by level 1 assets).

Chart 5

The funding mix of each bank is not only a major determinant of outflow, it also provides indications of the strength of each bank's funding. High portions of stable retail deposits and operational wholesale deposits indicate funding strength (see chart 6). Stable retail deposits are entirely covered by deposit insurance and either held in a transaction account or by a depositor that has other relationships with the bank.

Operational wholesale deposits are made up by unsecured wholesale funding or a collateralized deposit that is necessary for the bank to provide operational services as an independent third-party intermediary, agent, or administrator to the wholesale customer or counterparty providing the unsecured wholesale funding or collateralized deposit. Regions Financial and M&T report strong levels of those deposits as percentage of overall deposits (see chart 7). BB&T and Citigroup report a relatively low amount. JPMorgan Chase reports both a high level of wholesale deposits and a high percentage of those made up by operational deposits. BB&T reports the opposite.

Chart 6

Chart 7

At the median, retail deposits made up about two-thirds of the deposits of the banks subject to LCR, with slightly more than half of those deposits fully covered by FDIC insurance. Wholesale deposits accounted for the remaining third of deposits, and half of those were in operational accounts.

Most banks expect minimal cash inflows relative to cash outflows (11% at the median). However, the LCRs of banks that focus on credit cards (American Express and Synchrony Financial) benefit from the significant cash inflows resulting from minimum payments due on card loans and receivables. The LCR calculation also includes no outflows related to cancelable credit card lines. As mentioned earlier, banks with large broker-dealer operations also tend to have significant inflows related to secured lending they engage in.

Over Time, We Could Lower Ratings On Banks That Reduce Liquidity

The potential elimination or easing of the LCR for a number of banks doesn't necessarily mean they will respond with a significant reduction in liquidity. In some cases, the LCR may not be their binding constraint for liquidity management. That is, they may have internal liquidity stress tests and requirements that are more stringent than the LCR.

That said, if the Fed's proposal comes to fruition, it is possible that these banks could significantly reduce liquidity--perhaps closer to the levels banks without an LCR requirement often operate at. After all, holding excess liquidity does affect profitability, all else equal. While this is not our base-case expectation, over time, we could take negative rating actions on a bank that responds with substantially less conservative liquidity management, resulting in materially lower liquidity than peers taking into account outflow assumptions.