Could New Liquidity Requirements Sink the Ship?

Because the international banking regulations dubbed “Basel II” have not yet become operational in the US (and might not ever be), I was a little surprised, in recent weeks, to see two US fixed income analysts dissect the impact of still-in-the-works Basel III liquidity rules in recent weeks. At Barclays Capital Rajiv Setia, Anshul Pradhan and Amrut Nashikkar considered the potential effect of the “evolving paradigm” on bank balance sheets while RBS analyst Margaret Kerins looked at rules’ potential impact on demand for US agency securities. As a matter of fact, the Barclays analysts were surprised that the proposed “prescriptive liquidity regime” had thus far received so little attention in the US. Both reports gave me pause, so I thought I’d share them, along with other thoughts, here. The clarity with which both reports identified potential negative unintended consequences gave me pause, so I thought I’d discuss them here. Basel II Necessitates Basel III Basel II, as cognoscenti know, prescribes risk-based capital rules reflecting credit risk, operational risk and market risk in trading books (Pillar 1), principles of supervisory review (Pillar 2) and stronger disclosure of risk and capital profiles (Pillar 3). Of the three, the risk-based capital rules are the most extensively articulated, in two flavors, one an expanded matrix of Basel I weightings to incorporate risk rating and other factors, the other dependent on banks’ internal models for risk weightings. Regulators overseas were in the midst of implementing Basel II, and US regulators squabbling over it, when the financial crisis bubbled up. That crisis highlighted a number of shortcomings in Basel II – most notably it tended to lower capital just as more capital would have been wanted). So, beginning in 2008, the Basel Committee on Banking Supervision set out to address those shortcomings (bear in mind, the Committee is made up of the same regulators from the US and other G-20 nations, as well as 7 other jurisdictions with internationally active banks, that ultimately implement its “accords”). Their package of consultative proposals was delivered in December 2009 for comments ending April 16, 2010, with final rules anticipated by the end of this year and full implementation scheduled by the end of 2012. The enhancements that make up Basel III generally address three issues: capital, liquidity and “macroprudential” measures intended to offset cyclical capital pressures and systemic risk. The liquidity rules are intended to push banks to hold a higher percentage of assets in very liquid, lower yielding instruments and decrease reliance on wholesale sources of funds (such as repos and short term advances). In particular, the proposed liquidity rules prescribe two liquidity measures: • Liquidity Coverage Ratio (LCR) addresses assets that can be converted to cash to meet liquidity needs during an acute liquidity stress scenario specified by supervisors. LCR = institutions’ stock of unencumbered high quality liquid assets divided by projected net cash outflows over a 30-day time period. • LCR must equal or exceed 1.0 (e.g. qualifying assets must equal or exceed projected net cash outflows under stress). Moreover, 50% of the stock of highly liquid assets must be in cash or securities that carry a 0% risk weight under the Basel II standardized approach. • Stress scenario assumptions include a 3-notch credit rating downgrade, partial loss of deposits, loss of unsecured wholesale funding, reduction of term funding sources, significant increases in securities funding haircuts and increases collateral calls and calls on committed credit and liquidity facilities. • Net Stable Funding ratio (NSF) measures longer-term stable sources of funding relative to the liquidity profiles of the funded assets and as well as the potential for contingent calls on off-balance sheet commitments and obligations. NSF = available amount of stable funding (ASF) divided by required amount stable funding (RSF). • An institution’s ASF must exceed its RSF (ratio; 100%). • The Committee’s proposed rule weights the stability of funding sources, from 0% for wholesale funding, to 50% for wholesale funding from non-financial sources, 85% for retail deposits and 100% for term funding longer than one year and capital. • Likewise, RSF levels are defined for each asset class. An RSF factor of 0% is assigned to cash and securities maturing in less than one year, 5% to sovereigns with maturities of a year or more, 20% to corporates or covered bonds rated at least AA with maturities of a year or more, enjoying a deep, active and liquid market and a proven history as a liquidity source in stressed market environments 20%, and 100% to loans longer than a year. The proposed rules do not address the obligations of government-sponsored entities; implicitly US GSE debt and MBS would be treated like AA or better corporates. The Consultative Document stresses that these are minimum levels of liquidity for internationally active banks. As with the capital adequacy accord, “national authorities are free to adopt arrangements that set higher levels of minimum liquidity.” Effect on Agency Demand As RBS’ agency strategy analyst, Kerins naturally focused on the potential effect of these requirements on bank demand for agency debt and MBS. If adopted as currently written, she anticipates a shift in demand from agency securities to cash, Treasuries and other sovereigns. Indeed, she notes that shift could be substantial as US banks owned about $1.2 trillion in agency debt and MBS at the end of 2009 according to the Federal Reserve’s Flow of Funds (Z.1) report. Kerins also makes the intriguing general point: by narrowly defining the set of high quality liquid assets , the standard actually could reduce liquidity in the event of a systemic shock “as everyone is trying to sell the same assets”. However, given such undesirable unintended consequences, she expects the Committee to revise the rules. Alternatively, US regulators amended the Committee’s 2008 liquidity management and supervision principles to treat GSEs securities like Treasuries when they issued the 2010 Policy Statement on Funding and Liquidity Risk Management. Certainly, they can do the same when implementing liquidity measurement rules. Setia and his team at Barclays anticipate holdings of Treasuries and agency securities alike must increase, while loan portfolios must shrink (I’ll return to this point shortly). However, they note that holdings of GSE securities will be constrained both by their diminishing supply and by the requirement that at least half of the high quality assets in the LCR calculation have zero risk weights for regulatory capital calculations. (Ginnies have 0% risk weights, but Fannie and Freddie obligations have a 20% risk weighting.) Bank Credit Crimped The Barclays analysts performed a simple optimization exercise to demonstrate the dramatic transformation the proposed rules would wreak on bank balance sheets. Starting with FDIC call data as of year-end 2009 and the proposed rules as a base case and assuming two years to implementation, they quantify increases and reductions in assets and liabilities. (The rules are intended for internationally active banks, but using all banks is a reasonable first cut. The results are only meant to be illustrative.) In addition, they look a a variety of alternate scenarios, assuming lower leverage, wider funding spreads (on supply or downgrade), lower duration gap, equal treatment for agencies with Treasuries for liquidity purposes, and combinations of those scenarios. As of year-end 2009, the US banking system was far from being in compliance US banks. Exposed to a net cash outflow in 30 days of $2.8 trillion, it held about $1.9 trillion in unencumbered liquid assets under the proposed rules. That equates to total shortfall of around $850 billion. This does not take into account the fact that the Fed’s quantitative easing program has boosted cash in the system by almost $1 trillion. When (if) conditions finally warrant draining those reserves; the banking system’s cash position will shrink, its security holdings increase. Likewise, even if cash is invested in Treasuries or Ginnies, the shortfall would increase. In Barclays’ view, and I think this is a key point, under the proposed rules, the impact of draining reserves could be quite severe and have important implications for investment and growth. What about the stable funding requirement? Based on the haircuts in the proposed rules, the Barclays analysts estimates available stable funding will fall short of required by about $1.7 trillion or 13% of total US banking assets. Bear in mind this results reflects both the relative liquidity of assets and of funding. On the asset side, they observe that cash and cash equivalents (including excess reserves) significantly lower banks’ apparent stable funding requirements. Furthermore, just 10% of assets are in highly liquid instruments, like Treasuries or agency debt and MBS, with a relatively low RSF factor of 10%, while 60% of assets are in loans and less liquid securities with high RSF factors necessitating higher stable funding. In other words, assuming no change in funding banks must let loans roll off in favor of larger concentrations of highly liquid, highest quality securities. What about the funding side, then? Barclays observes that banks still rely on over $1 trillion (about 7.5% of liabilities plus capital) in wholesale funding sources. Those liabilities that do contribute to stable funding are retail deposits (50% of sources of funds) and equity/long term debts (18%). Wrestling Balance Sheets Into Compliance It’s not news that, with the crisis and “Great Recession”, loan demand has fallen and lending requirements tightened, leading banks to make fewer loans and to allow loan portfolios to shrink. Total assets are roughly flat, with the runoff redeployed into securities. At the same time, banks have increased retail deposits, shored up capital and decreased their reliance on short term funding. The net effect has been to move banks closer to Basel III targets: Barclays estimates that the NSF shortfall has declined from roughly $3 trillion at the start of 2008 to $1.7 billion as of December 2009. However, these are strategies that banks arguably would have followed without the prospect of new liquidity rules. What happens to asset and liability mixes if the Basel III liquidity rules go into effect? Barclays’ optimization exercise – optimizing profits given constraints of the rules – indicates that Treasury holdings could jump by about $900 billion, agency debt and MBS by about $300 billion, while loans outstanding must fall by about $600 billion. On the liability side, wholesale funding would be virtually eliminated and the percentage of retail deposits forced to grow. Higher priced funding and lower yielding assets means margins would shrink. Indeed, bank profitability is lower in every scenario examined. In other words, a dispassionate look at the proposed liquidity rules suggests two troubling outcomes. One would be a reduction in bank lending, the other a relative disincentive to would-be suppliers of bank equity and long term funding. Furthermore, a reduction in bank lending could occur just as economic growth – and loan demand – pick up. Setia and his team constructed a simple feed back model of the interplay between bank lending and economic growth (based on historical patterns). Not at all surprisingly, they find that the rules will impose a substantial drag on growth, a drag that could be yet more significant in the Eurozone, where “bank loans constitute a much bigger component of the overall supply of credit”. NOTE: Linda Lowell writes a regular column, called Kitchen Sink, for HousingWire magazine. Editor’s note: Linda Lowell is a 20-year-plus veteran of MBS and ABS research at a handful of Wall Street firms. She is currently principal of OffStreet Research LLC.