Too Big To Fail

#45: On Banks, Bonds, and Bailouts.

“If the crisis has a single lesson, it is that the too-big-to-fail problem must be solved.”

-Ben Bernanke, September 2, 2010

Back in the mid 2000’s, there was a big problem with banks. Over the course of decades, leverage in the banking sector grew tremendously. While banks reported reasonable equity capital positions, the actual cash held by U.S. Commercial Banks had barely budged, even as liabilities grew year after year.

In 1982, cash leverage of the banking sector stood at 10:1 - for every $10 dollars of bank liabilities (primarily customer deposits), banks held $1 of liquid cash. By 2008, this leverage metric had tripled to 33:1¹.

The shrinking proportion of liquid assets was not a problem during the good times. Bank lending continued to grow the broad money supply, and without pressure of credit losses or counterparty concerns the deteriorating liquidity position of the banking sector remained in the shadows. It was only once banks were forced to absorb losses (primarily emanating from the subprime mortgage crisis) that the weak capital structure of the banking sector was revealed.

Initially, the Federal Reserve was content to let the market resolve the issue, but a wave of bank consolidation and emergency buyouts ultimately gave way to outright failures. Concern gave way to contagion and panic. With the financial sector in turmoil, the Fed and Congress relented.

To address a burgeoning global financial crisis, banks needed to be liquified, fast. The primary purpose of the Emergency Economic Stabilization Act of 2008 (including the Troubled Asset Relief Program) was to replace bank’s illiquid assets like Mortgage-Backed Securities (MBS) with the liquid asset - cash.

By January 2009, thanks to the coordinated liquidity infusion (or bailout), U.S. bank leverage fell back to 11:1 - resetting the prior two decades of expanding cash leverage in the course of three months.

The bailout, while effective, was not heralded as an accomplishment, but rather as the best of bad options - a necessity of the moment, or a deal with the devil. Nor was the bailout of Wall Street well received by a population entering a deep recession. Even the architect, Ben Bernanke, made it clear that a too-big-to-fail precedent was untenable in the long run, citing moral hazard and the preferential treatment of large, entrenched institutions.

In his words, “the too-big-to-fail problem must be solved… Few governments will accept devastating economic costs if a rescue can be conducted at a lesser cost; even if one Administration refrained from rescuing a large, complex firm, market participants would believe that others might not refrain in the future.”

Over the coming years, a slew of regulations from Dodd-Frank to Basel III were implemented to better control bank leverage, ensure adequate liquidity, and avoid another banking crisis.

Today, there is little risk of a repeat widespread liquidity crisis at large U.S. banks² - new regulation, and more importantly, a decade of QE, mean that major banks have little risk of actually running out of money. In other words, while the underlying “too-big-to-fail” question has never been addressed, it is no longer a practical concern for systemically important banks because of the firehouse of cash pumped into banks since September 2008. Aggregate cash leverage in the bank sector today sits below 7:1, nearly five times lower than August 2008.

Yet, despite the cash piles sitting at money-center U.S. banks, the financial system remains prone to dysfunction. Tighter regulation has moved risk-taking off banking balance sheets and onto the broader non-bank institutions and the market-traded securities. Risk hasn’t been eliminated, it has been transferred. The problem of too-big-to-fail has not been eliminated, it merely exists on a different rung of the financial ladder.

Banks and Bonds

If you squint, banks and bonds aren’t all that different.

A bank sits between lenders and borrowers. On one side, depositors hand money to a bank and receive a piece of paper in return guaranteeing access to that money in the future. Loan underwriters, portfolio managers, and risk officers at the bank determine how to lend money to borrowers, while ensuring an optimal mix of illiquid and liquid assets to satisfy cash flow needs and potential withdrawal demands of depositors. The bank earns the difference between its lending rate and the interest it pays to depositors - known as a net interest margin (NIM).

Traded debt markets - bond markets - connect lenders to borrowers directly. Direct lending via bonds provides higher returns than bank deposits by eliminating the middleman, but also puts the burden of credit underwriting and liquidity management directly on the lender (and out of the purview of bank regulations). Bonds are illiquid - the lender can’t demand cash back from the borrower until maturity. Instead bondholders rely on an active trading market for liquidity.

In the same way that banks depend on adequate balance sheet cash to avoid bank runs and failures, bond markets are dependent on the trading liquidity, as measured by average daily trading volumes or book-depth. And while a bank’s balance sheet is managed by its employees, the bond market uses price to mechanically balance supply and demand.

Both banks and bond markets are large, mostly illiquid, pools of debt. And both are subject to runs.

If investors suddenly rush to sell bonds to raise cash, it will drive down the price of those bonds to induce more buyers. Of course, this lowers the value of the bond not just for the sellers, but the entire universe of bonds outstanding. If trading liquidity is unavailable, the entire pool of debt can be subject to massive drops in value, with cascading effects to investors and throughout the financial system (i.e., non-bank solvency, equity market crashes, extreme cross-asset volatility, and gapping credit spreads).

The bond market is much larger than the combined size of commercial banks. The value of just U.S. Treasuries held by the public, a subset of the entire federal debt, today stands at $24 trillion - greater than the $22 trillion of combined banking assets. In total, the Securities Industry and Financial Markets Association (SIFMA) estimates the total size of the U.S. bond market at $46 trillion in 2021, which includes government debt, corporate bonds, and asset-backed securities like MBS. While one could argue that banks are still dollar-for-dollar more systemically important than bonds, a far greater portion of investor wealth and financial assets exist as bonds.

There were no bank runs in March of 2020 as the COVID panic set in. Instead, there was a massive dash for cash in financial markets - a bond run. The results were financial dysfunction only surpassed in recent history by the global financial crisis.

Faced with a new crisis, the Federal Reserve was again confronted with whether to let the private market participants to price risk and accept loss, or step in and backstop the market.

Having already fired up the QE printer by late February (which in twelve short years had transformed from a crisis intervention to a run-of-the-mill monetary policy), the Fed extended its liquidity support far beyond U.S. Treasury and agency-backed MBS. On March 23, 2020 the Fed announced unprecedented support, backstopping foreign exchange, corporate debt, commercial paper, money markets and beyond, far eclipsing the scale of intervention in 2008. All of these measures socialized the risk of financial institutions and investors, but ultimately were successful in restoring order to the bond market.

Yet unlike a decade ago, there have been no caveats or warnings, no “best of bad choices” speech. Not only have we not solved the problem, we no longer believe one exists.

Conclusions

In yesteryear, the Fed acted as a lender of last resort to the banking sector via the discount window. In 2008, the Fed took extreme measures to bail out banks, even as it recognized the hazard of “too-big-to-fail”. By 2020, there were few corners of the debt markets that were considered small enough to fail.

Somewhere along the line, the bailouts became business-as-usual. Both the Fed and the broader public seem far more accepting of the idea that any disruption in the financial system should be met with greater support for the monetary authority.

Maybe the fact that COVID generosity was also extended to everyday folks in addition to financial markets made it more palatable and democratic than the GFC bailouts. A booming stock market certainly didn't hurt. Or maybe it was the notion that COVID was a unique exogenous shock rather than a self-inflicted wound. Or perhaps we've simply come to accept our role as the damsel, waiting for our white knight. Or, maybe Bernanke was just wrong.

I remain skeptical. While the expedient, and now expected, solution is for immediate intervention, it’s hard to see how suppressing price discovery leads to more efficient markets over the long run. Central bank intervention merely allows investment returns to be privatized while investment risk is socialized - a deeply regressive tax in the form of dollar dilution and dwindling spending power that benefits the richest among us.

If the end goal is to build a financial system that is resilient to shocks, we might just have to let it face one. My guess is we’d come out stronger on the other side.

---

If you enjoy The Last Bear Standing, please subscribe, hit “like”, and tell a friend! Let me know your thoughts in the comments - I respond to all of them.

As always, thank you for reading.

-TLBS

1. Reserve balances (cash which private banks held at the Fed), stood at only $10 billion, making up a tiny fraction of cash assets or total assets. Today that figure sits at $2,998 billion.

2. This is perhaps a controversial statement in light of Silvergate Bank liquidating and an active bank run on Silicon Valley Bank - two banks with unique exposure to cryptocurrency and the venture capital market.

More generally speaking, here are several very important caveats when thinking about.

First, we must differentiate between solvency - the equity capital of a bank - and liquidity. It is very possible that equity holders of a given bank may incur substantial losses or total wipeout due to devaluation of the bank’s assets, particularly as rising interest rates reduce the value of debt securities held on balance sheet. It is less likely, in my view, that these banks will become illiquid (run out of money), which would be a far greater systemic threat.

For the record, while total unrecognized losses in held-to-maturity bank portfolios today stands at around $300 billion, the total equity capital in the banking sector is $2,000 billion (which already incorporates recognized but unrealized losses in available-for-sale portfolios).

Second, we should differentiate between large and small banks. While large banks continue to hold meaningfully higher cash balances as a percent of total assets than they did in 2019, small banks have seen much greater cash outflows over the past year and are in roughly the same position as 2019. Therefore, liquidity stress is much more likely to surface in these small banks. While certainly not a good thing, it is far less concerning than 2008 when large money-center banks like JPM, WFC, BAC, C, etc. were all under significant liquidity strain.

My greatest concern relating to bank liquidity is not that banks run out of money, but rather than they reach their minimum reserve requirements and stop lending to non-bank institutions, as was the case in the 2019 repo market fiasco. This is the great irony of stringent bank regulations. By forcing banks to hold a substantial level of reserves at all times, you can ensure that banks don’t run out of money - but also trap a lot of money in those banks. The liquidity risk is transferred from banks to those who rely on banks for funding.

It is also possible that I’m wrong. Perhaps the degree of interest-rate induced credit losses combined with draining bank liquidity via QT will lead to the failure of more important banks, or that the failure of smaller banks could cascade via contagion to larger ones, and funding markets freeze. But until we see greater evidence of this, my concern remains outside the walls of commercial banks.