The Federal Reserve's Coronavirus Crisis Actions, Explained (Part 2)

Yesterday I wrote about the Federal Reserve Board’s announced domestic actions taken Between March 9th and March 20th to respond to the Coronavirus-induced depression. Today I’ll be covering the rest of the domestic actions the Federal Reserve has taken up until this morning. Tomorrow I’ll be covering their international actions over this time period. In a sense though, this post is far simpler than Part 1. Part 1 covered a series of actions over an extended period. Part 2’s centerpiece is one announcement and the rest are simply details.

First though, I want to cover two critical things I skipped over yesterday (that post was already far too long). In what was announced March 15th, the Fed did two seemingly minor things that are actually very important. The first is it encouraged banks to take out “daylight overdrafts” from the Federal Reserve regardless of whether they were collateralized or not. This seemingly minor change is actually critically important. Much of the discussion of “Repo Madness” last year revolved around how the new regulations combined with the Fed’s new operating procedures creates periodic shortages of “intraday” liquidity meaning liquidity you need basically immediately to settle payments. As Zoltan Poszar put it, switching from a credit-based clearing system to a token-based clearing system makes having pre-existing holdings of tokens (settlement balances) far more important and valuable. In some ways, this is intuitive. If you could run an overdraft with your bank at any time to make any payment and only had to “get back to zero” every 30 days, you’d probably keep less in your checking account as well. When the Fed has trouble supplying sufficient tokens for the system, it periodically freezes up. The Federal Reserve encouraging intraday credit and disclaiming any barrier to intraday credit is essentially going back to a clearing-based system. This change is good, though it makes one wonder- if daylight overdrafts are best in a crisis, maybe they’re just best all the time?

The second change I skipped over is in some sense trivial and, in another sense, absolutely gigantic. It is worth quoting in full:

For many years, reserve requirements played a central role in the implementation of monetary policy by creating a stable demand for reserves. In January 2019, the FOMC announced its intention to implement monetary policy in an ample reserves regime. Reserve requirements do not play a significant role in this operating framework.

In light of the shift to an ample reserves regime, the Board has reduced reserve requirement ratios to zero percent effective on March 26, the beginning of the next reserve maintenance period. This action eliminates reserve requirements for thousands of depository institutions and will help to support lending to households and businesses.

Let’s break this down sentence by sentence. The first sentence explains something that most monetary policy technicians have known but the broader public generally has not- requiring banks to hold a certain quantity of settlement balances in proportion to their liabilities doesn’t mechanically reduce lending by preventing banks “lending out” settlement balances to retail and corporate customers. There is no “Money multiplier” by which total bank credit is limited by a certain pre-existing quantity of settlement balances. Instead, the purpose of “reserve” requirements was to, as the announcement said, stabilize the demand for settlement balances. By having a stable, predictable, demand for settlement balances the Federal Reserve was able to inject exactly how many settlement balances the banking system needed. Creating a stable demand for settlement balances became far less important after 2008 (but before the latest round of Basel regulations were implemented) since the quantity of settlement balances outstanding far exceeded “reserve” requirements. With the new Basel regulations, “reserve” requirements were superseded by more stringent liquidity regulations that created a far larger demand for settlement balances. Reserve requirements lost their operating purpose. Not only that but, as it says, in this environment they’ve decided to aim for keeping the banking system filled with “unencumbered” settlement balances.

Thus, this action is mainly cleaning up a superfluous regulation. The one advantage it does have is that it frees up the balance sheet of smaller banks which aren’t subject to the Liquidity Coverage Ratios I discussed in Part 1. Reserve requirements may not prevent bank lending in the traditional sense of reducing their ability to borrow, but they are a non-reciprocal obligation which require holding a certain quantity of a low yielding asset. That balance sheet space can take up higher yielding assets and more generally lead to higher profits (which, if retained, could sustain more lending from a capital perspective over the longer term). More than anything else though, I think this change is padding to make it seem like they are pulling out “all the stops” to support credit markets.

Now that those are dealt with, we can finally move on to this week.

March 23rd

The main statement came first thing in the morning on Monday and reads like the Federal Reserve putting all of its cards on the table. It once again puts more liquidity in government and government guaranteed securities markets to offset the effect of markets dysfunctioning under the strain of the crisis. One small way you can tell how the urgency has changed is simply by how the language describing circumstances has evolved.

March 23rd:

The coronavirus pandemic is causing tremendous hardship across the United States and around the world. Our nation's first priority is to care for those afflicted and to limit the further spread of the virus. While great uncertainty remains, it has become clear that our economy will face severe disruptions

March 15th:

The coronavirus outbreak has harmed communities and disrupted economic activity in many countries, including the United States. Global financial conditions have also been significantly affected.

March 9th:

Federal financial institution regulators and state regulators today encouraged financial institutions to meet the financial needs of customers and members affected by the coronavirus. The agencies recognize the potential impact of the coronavirus on the customers, members, and operations of many financial institutions and will provide appropriate regulatory assistance to affected institutions subject to their supervision.

In less than two weeks the Federal Reserve moved from boilerplate regional natural disaster language to probably as disturbed and concerned as bureaucratic press releases can get. Another way to tell how seriously the Federal Reserve is taking a crisis is by how many new acronyms they’re announcing in any one statement and how much the old acronyms are expanding. By my count this statement introduces three new acronyms to this crisis and expands the two existing ones. That’s Fedspeak for running around with your hair on fire. To break this down, let’s start with the “old” acronyms.

The Money Market Mutual Fund Liquidity Facility is the first up. Apparently MMMMLF has too many Ms in it and just looks silly so the facility is acronymed MMLF. You can say a lot of things about the Federal Reserve, but you can’t say their acronyms don’t have a lot of thought put into them. Anyway, this change is relatively minor as things go. They’re allowing purchases of bank negotiable certificates of deposit to prevent bank funding rates in this market from spiking. Presumably there is worry that as non-financial corporations draw down their liquid assets, markets like this which are presumably otherwise safe (but not completely safe as large denomination NCDs are not FDIC insured) will come under strain. The second change opens up the MMLF (see it just rolls off the tongue now) to Municipal Variable Rate Demand Notes (VRDNs) that you can “cash in” at any time. The logic clearly is these securities have many properties of short term borrowing, so they can effectively be treated as short term borrowing.

The Commercial Paper Funding Facility was also expanded, while its interest rate was also cut. Originally it was priced at the 3-month Overnight Index Swap (OIS) +2% whereas now it is priced at OIS +1.1%. The Overnight Index Swap is a derivative contract which allows market actors to “hedge” or bet on what short term borrowing rates will be over the course of the contract.  The premium a lender sets over OIS is basically what they’re charging to be compensated for the “credit uncertainty” they are taking on. The second change adjusts for worsening credit conditions by “grandfathering in” corporate securities that were at the minimum eligible rating on March 17th but subsequently lost that rating in the last 9 days. It would defeat the purpose of the facility if everyone dropped out of it because of the impact of worsening economic conditions on their balance sheets.

Now onto the “new” facilities. The first we’ll start with is another star of 2008, the Term Asset-Backed Securities Loan Facility (TALF). In the Great Financial Crisis of 2007-2009, TALF was announced November 28th, 2008. Then TALF only had a handful of eligible assets. “The underlying credit exposures of eligible ABS initially must be auto loans, student loans, credit card loans, or small business loans guaranteed by the U.S. Small Business Administration”. Coronavirus’s TALF in contrast applies to nearly any household and small business credit exposure you can imagine:

1) Auto loans and leases;

2) Student loans;

3) Credit card receivables (both consumer and corporate);

4) Equipment loans;

5) Floorplan loans;

6) Insurance premium finance loans;

7) Certain small business loans that are guaranteed by the Small Business Administration; or

8) Eligible servicing advance receivables.

As extensive as this seems, it is even more extensive than 2008 in a way that interacts with the “stimulus” bill passed last night- the number of small business loans guaranteed by the SBA will be far greater this time around. The only way in which this TALF is less expansive than 2008 TALF is in who qualifies for it. In 2008 “All U.S. persons that own eligible collateral may participate in the TALF”. This time around “All U.S. companies that own eligible collateral and maintain an account relationship with a primary dealer are eligible to borrow under the TALF”. It’s not immediately clear to me why they’ve narrowed the eligibility in this way, though I suspect it is operationally simpler to administer the program through customers of primary dealers than any entity or person who walks off the street. Finally, these loans are non-recourse, like some of the loans discussed yesterday, which makes TALF more like a purchase facility (though these purchases still take up balance sheet space).

This leaves us the final two facilities which are very similar: The Primary Market Corporate Credit Facility and the Secondary Market Corporate Credit Facility. These brand new facilities are really the defining Federal Reserve intervention of this crisis so far. Other interventions have been either standard, fixes to longstanding technical problems or expansions of facilities we saw in 2008. These are completely new to the United States and are effectively multi-trillion dollar corporate bond purchases. In another sense though, they’re a legally innovative version of what other central banks around the world have been trying since their interest rates went to zero (or negative). Can’t lower the risk free rate? Well lower corporate spreads over the risk free rate. The major difference here is that rather than trying to suppress spreads below their typical position, this facility is trying to keep spreads where they were before Coronavirus broke out and incomes for both businesses and households collapsed across the board.

The main difference between these two facilities is that the Primary Market Corporate Credit Facility (PMCCF) purchases securities directly from corporate debtors while the Secondary Market Corporate Credit Facility (SMCCF) purchases on the open market. The SMCCF is also authorized to purchase Exchange Traded Funds, presumably on the theory that inflows into these Funds will also be effective at easing credit conditions since they offset the outflows coming from distressed or panicked investors. Eligible securities must be rated at least BBB-/Baa3 by a “recognized” rating agency. This rating is the lowest tier of “investment grade”, in other words the “meh” of investment grades. You can think of this rating as like a “C” you’d get in school.  The PMCCF is limited to maturities that are 4 years or less while the SMCCF is limited to maturities that are 5 years or less.

The caps to the programs are also interesting. The Secondary Market Corporate Credit Facility is capped to purchasing a maximum of just 10% of any individual company’s peak outstanding securities in the last year. So, for example, if Apple had a peak amount of 300 Billion dollars of outstanding securities in April 17th 2019, the maximum amount of Apple corporate securities the SMCFF could hold was 30 billion dollars. The Primary Market Corporate Credit Facility in contrast is capped to purchasing a maximum amount of securities equal to over 100% of the peak outstanding securities AND loans in the past year. The better the investment grade, the higher the cap is in this facility. These very high caps reflect the fact that non-financial corporations have extraordinary credit needs at a time where sales and thus revenue has utterly collapsed.  The motivation behind the differential caps based on investment grade rating seems to be that the ideal participant is a previously extremely creditworthy corporation which was hit extremely hard by the Coronavirus-induced depression and needs maximal support as it runs down its credit lines to make payroll.

It is important to know that these facilities both exclude from the program corporations which are getting direct financial assistance from the legislation passed yesterday. Thus, there will not be “double dipping” between these facilities and the “stimulus” bill. What’s interesting about this is it implicitly suggests the Federal Reserve expects markets to keep borrowing costs down for corporations that get direct assistance from congress. It will be important to watch whether that plays out as expected or whether these facilities will be revised to include those industries. The announced facilities say nothing about “4 trillion in loans”, the number bandied about in the media.

Given that both facilities are capitalized by the Exchange Stabilization Fund to the tune of just 10 billion dollars it is likely these programs that are in large part what congress and the administration has in mind when it focuses so tightly on allocating 454 billion dollars to the Federal Reserve’s lending programs. This further suggests that there is no significant quantity of capitalization the Fed needs to launch these programs since the programs have been announced and there’s no operational- or even suggested- limit to how many purchases these facilities can engage in without “further capitalization”. The arbitrariness of the numbers involved suggests that it is the sturdy rule of the “ten to one ratio” in undergraduate money and banking textbooks which is motivating the specific quantity of 454 billion dollars since it would be a tenth of a 4.54 trillion dollar purchase and lending program.

The second announcement on March 23rd, in comparison to the first was tiny. Yet, it is significant. It is a new rule that follows up on the March 17th announcement that revises the definition of “eligible retained income” by updating another part of our post-crisis capital regulations to use the definition. In this case, what they are updating is the “Total Loss Absorbing Capital” (TLAC) rule which is similar to capital requirements but broader in that it includes debt with extended maturities along with equity that can be wiped out in extreme circumstances. The purpose is the same as yesterday, allowing banks to “use” their buffers to support credit to households and banks while steadily reducing dividends, buybacks and bonuses rather than having to suspend them.It remains bad policy that paying out dividends, buybacks and bonuses is tied to maintaining lending and they should be immediately suspended across the board.

March 24th

The first announcement on March 24th formalized last week’s announcement about daylight overdrafts. There has been ongoing rulemaking over the maximum amount of “daylight credit” a financial institution could receive from the Federal Reserve. These new rules were scheduled to be implemented on April 1st but “out of an abundance of caution” they are being delayed until October 1st. While making sure intraday liquidity is always guaranteed during this crisis is a good thing, it is concerning that they are still planning on moving forward with this rule. If its true that “The availability of intraday credit from the Federal Reserve Banks supports the smooth functioning of payment systems and the settlement and clearing of transactions across a range of credit markets”, it seems like limitations on intraday credit are dangerous. If its good enough for a crisis, its good enough in advance of a crisis.

Coincidentally, the final announcement on March 24th is in essence a heightened form of the announcement on March 9th a little over 2 weeks earlier. It formally moves regulatory priorities from bank examination and responding to “non-critical supervisory filings” towards monitoring and outreach related to Coronavirus. Large banks still need to submit their Comprehensive Capital Analysis and Review (CCAR) by April 6th but aside from this, normal regulatory activity has been temporarily eased. It also encourages banks to “work with” debtors who can’t make payments. Specifically, loan modifications during Coronavirus will not be treated as “troubled debt restructurings”. In plain English this means that if you give a debtor a temporary break on their debt, that loan won’t be reclassified as higher credit risk or relying recognizing losses which would negatively affect a bank’s capital position. This is good, but won’t be nearly enough to give households and businesses the debt relief they need.

Conclusion

This brings us up to date. There were 2 announcements today but they are pretty minor and similar to the final announcement of March 24th. I’d summarize this flurry of action this way- regulatorily they finalized easing measures that were announced in more general terms last week while expanding their legal engineering to create programs for broad-based corporate bond buying (while expanding other programs at the edges to support corporate credit markets and marginally help municipal and state debt markets). While this intervention is sweeping and unprecedented in the United States, it will also not be enough. Corporations don’t need low cost debt, they need revenue to make payroll.

Given that nearly 500 billion dollars of the announced $2.2 Trillion dollar “stimulus” has gone to supporting this program- and there may be far more loan programs in the legislation- it doesn’t seem like very much income is forthcoming for households or businesses. Unless there are plans to make corporate america whole after the fact, I think it can be overstated how extensive the support that the business sector is getting even as households get much less. I’m also concerned that the large quantities of corporate debt buying will suffer what conventional quantitative easing suffers from- purchase quantities don’t have the impact on interest rates that we hope or expect of them. Rate setting is far better, more efficient and honest.

-Stay Safe,

Nathan Tankus