Silicon Valley Blues

by
Peter Radford
My articles My siteMy booksMy videos
Follow on:

from Peter Radford

The circumstances of Silicon Valley Bank are well rehearsed by now.   The bank sat at the epicenter of the tech-bro start-up ecosystem and played a pivotal role in the collection and disbursement of all the cash that flows through that system.  It was an extremely odd bank.  Unlike the everyday banks most of us deal with it had very few deposits that originated from regular customers.  Most of its deposit base consisted of the chunky piles of cash belonging to start-ups and their various hangers-on.  That meant the average size of each deposit account was well in excess of the limit the FDIC insures.

This odd customer base added to the strain on the bank during recent years when the combination of low interest rates and excess cash slopping about the economy led to a surge in start-up activity.  The bank benefitted from that activity — if benefit is the right word — and saw its deposits grow rapidly.

This presented the bank with a problem.  What to do with all those incoming deposits?   Because the bank had so few ordinary customers, and generated so few regular assets, it had few options.  Unwisely it chose the riskiest:  it bought long dated US treasury bonds in order to eke out the extra yield that is associated with such long dated maturities.  This is all well and good in a low interest rate environment.  Those bonds are usually thought of as being both safe and liquid — meaning that the bank can sell them easily to generate cash to pay any depositor looking to withdraw their deposits.

But as the Fed drove interest rates up to combat inflation the bank found itself squeezed in two ways.

First, the tech funding boom began to falter.  The inflow of new cash slowed and began to reverse.  Those start-ups began drawing down on their deposits.  The bank needed liquidity to meet those withdrawals.

Second, rising interest rates made those Treasury bonds lose value.  Bonds with an interest rate of 2% are worth less when rates have risen to 4%.  They begin to trade at a loss.  So when SVB started selling bonds to meet the withdrawal demands of its customers it simultaneously started to lose money.  Lots of money.

Even this would have been manageable as long as the rate of withdrawals was more a trickle than a flood.  Yes, the losses would have impaired the bank’s capital which it would have needed to replenish.  But if that had been done in an orderly fashion the bank may have survived.

This is where the oddity of the bank became its weakness.  The tech ecosystem is a tight knit one.  As rumors began to leak out that SVB was looking to raise capital, alarm bells went off. Social media lit up with urgent warnings.  Clever people of all sorts of persuasions started to tell their friends to pull their deposits out.  Clearly, they all argued, the bank was in trouble.  It was better to be the first rather than the last to get your cash out.  Their calls for action exacerbated the very crisis they were warning of.

So a classic bank run began and the bank was doomed.

The Fed had to step in and give its backing to the depositors.  The question was to what degree that backing ought be.  Should the Fed stop at the FDIC insurance limit of $250,000?  Or should it go beyond, as it has in the past, and back the entire deposit base?

The Fed chose the latter, more conservative, approach.  All the bank’s depositors will be made whole.  Those Silicon Valley libertarians have been saved by the government.  What irony.

The panic within the tech ecosystem at the end of last week manifested itself in all sorts of desperate and unseemly calls for government bail outs.  Apparently oblivious to the obvious contradiction between their philosophy of individual grit and liberty — the stuff of the myth surrounding our obsession with “founders” and “disruptors” — and the urgency of the call for the government to protect that self-same start-up culture, literally thousands of participants in the ecosystem worked the phones to lobby for a bail out.

It worked.

Silicon Valley is now functioning happily in large measure because the Fed saw fit to extend its support beyond the mandated FDIC limit.

Now come the cries of “moral hazard”

Every time the government bails out a bank, we are told, it is sending a message to the remaining banks that they need not be too careful.  They will survive no matter what.  The Fed will backstop them.  They can play risky games to eke out extra profit for their shareholders and will never have to pay attention to the downside.  Where, once, this moral hazard extended only to those banks deemed so large that their collapse would endanger the entire financial system, it now appears the backstop has been extended down to smaller banks.  Is the entire system now protected beyond the FDIC insurance limit?  If so, how does this affect that way in which banks will behave?

There is no easy answer.  Banks are vital to the functioning of our modern economies.  They are deeply embedded in everything that goes on in the day-to-day economic activity we all take for granted.  Simply allowing the odd bank to collapse in order to avoid introducing moral hazard appears to be an easy policy.  The problem is that history suggests that this obvious answer is no answer at all.  Once one bank goes down panic stirs amongst depositors of other banks and even the strongest of banks can get swamped by a tsunami of withdrawals.  Banking is not a cluster of individual banks.  It is a system.  More to the point, it is a system built on the flimsiest of assets: public trust.  Once trust goes, they system crumbles.

So eliminating moral hazard is itself hazardous.  Banking is inherently unstable.  The essence of banking is to take risks.  Sometimes those risks will destroy the bank.  How are we to allow this natural process to unfold without it spreading panic and dysfunction beyond the bank in question?  It is always better public policy to err on the side of caution and save the depositors of a failing bank from loss.  That stems the decline in trust and the system stays afloat.

But isn’t this dangerous?  How do we punish poor bank management?  Isn’t that the essence of good old capitalism — to let banks fail is to instruct the survivors to behave in better ways.

And here we get lost in the thickets of semantics.

If we “bail out” the depositors are we really creating moral hazard?

Not if the shareholders, other investors in the bank, and its managers all lose.  They are not saved.  They all lose their money or lose their jobs.  That’s a pretty clearcut message to everyone else.  So the issue of moral hazard is avoided with respect to ownership and management.

So what about the depositors?  Ought they, too, not be punished for not being sensible and thinking about the risks they were taking by putting all their money into one bank?  Should they not lose too?

In typical bank failures of the past, uninsured depositors — those with balances of over the FDIC limit of $250,000 — have experienced a loss of around 15% of whatever their exposure was.  They don’t lose everything because even a rotten bank still likely has assets that can be sold off to raise funds to pay depositors.  That loss is not catastrophic, but it can be really inconvenient, particularly as the liquidation of assets can take a while.

From the regulator’s perspective even a rapid liquidation process represents a window of doubt.  They much prefer a more rapid solution in order to calm the banking system and reassure depositors as quickly as they can.  So the best option becomes one in which the regulator seizes control of the failing bank and then arranges its sale to another, healthy, and usually larger bank.  Typically this takes place over the course of a weekend.  The failed bank is closed on Friday, and its branches re-open the next Monday under the colors of a new bank.  The purchasing bank buys the assets of the failed bank and honors the withdrawals of its depositors as if nothing had happened.  The inconvenience is compressed into one weekend during which the depositors of the failed bank would not be able to make withdrawals.

This tried and true solution could not be put in place in SVB’s case because its book of business was so peculiar that the potential purchasing banks expressed concern about the assets they were being asked to buy.

So when SVB re-opened the Monday after its failure it was under a new, temporary, name and under the day-to-day management of the FDIC.  The Feds keep a roster of experienced and trusted managers who can be called into run such state-run banks whilst a longer term solution — usually the piecemeal sale of assets — is undertaken.

But back to the issue of moral hazard.  The commentariat is full of discussion this week about how the SVB wind-up affects future banking practices.  Are we now in an era of total deposit insurance?  Is that $250,000 limit now redundant?  Have we made the banking sputum more reckless?

The answer is not clear.  And that has to do with the nature of our banks.

The problem is that the business of banking is not homogeneous.  Banks are bundles of different businesses or activities.

Some of these activities are what we call public goods — we all need a properly functioning payment system.  We need to feel safe depositing our cash into an institution that knows how to account for, manage, and move those deposits whenever we wish.  Why should any of us have to go through long winded due diligence just to deposit our cash?  Perhaps a large multi-national corporation needs to do so, indeed we should expect them to, but they all have complex cash management arrangements precisely because of this.  They protect themselves well.  If they lose money because they messed up their cash management, they do not deserve to be helped by a bail out.

The rest of us, including small businesses and those start-ups in Silicon Valley, are not that sophisticated.  We cannot be expected to spend time and effort executing complex cash management strategies.  We look at banks as utilities providing simple, safe, and reliable services.  Banks in this regard are giant accounting functions.  They are operational.  They need an eye for meticulous detail and error avoidance.

This is a very different activity from the riskier business of assessing the creditworthiness of potential borrowers, allocating funds accordingly, and then managing the various disparate cashflows such allocation inevitably creates.  Risk taking both in calculating whether someone is creditworthy and in matching the maturity of the incoming and outgoing cashflows is a skill banks specialize in.  They often get it wrong.  Their world is a very uncertain place.  And when they make mistakes, like Silicon Valley Bank, their very existence is threatened.  It is this aspect of banking that has, historically, been the cause of bank failures. — once word gets out that a bank has made bad bets, its depositors try to get their money out quickly.

Bank managers regularly make wrong or foolish decisions with regard to risk.  They misjudge the effects of interest rate changes.  They misjudge the ability of their customers to repay loans.  And so on.  Risk management is a central and, well, risky activity.  No one is perfect.  So banks fail.  The managers and investors in such risky activity need to pay the price when they get things wrong.  They certainly reap the reward when they get things right.

Every economy needs these two activities.  It needs a safe payment system.  It needs someone to take credit and balance sheet risks.  But we muddle the two.  Our banks do both.

So when the risk of the second threatens the safety of the first the Fed is always forced to act.  The functioning of the public good takes priority over the disciplining of the private sector activity.  Because we need our cash to flow easily and on demand, we protect the risk activities as well as the accounting activities.  We have no choice because our banks are a muddle of both sorts of banking.

If we want to protect depositors fully, no matter what their balance, and yet we want also to punish management for making poor risk decisions,  then we need to address and sort out this muddle.

Our perpetual inability to engage in sensible conversations about the role of the government in the economy — we have been crippled in that regard for decades since the stupidity of neoliberalism took hold of our political elite in the 1970s and 1980s — prevents us from simplifying policy.  And, in this instance, inevitably gets us into these debates about moral hazard.

Private sector banking is prone to crisis and loss.  Public attitudes towards private banking ought center on our need for credit allocation and the provision of funds to the real economy.  We ought assure that the flow of funds is reliable and is not threatened by panics.  Bankers, though, will make mistakes.  As long as we think the allocation of credit should be a private sector activity we ought allow those mistakes to cost the bankers and their investors exclusively.  Perhaps they will make fewer mistakes if they know there is no cavalry coming to their rescue.  Perhaps competition will sharpen and improve the flow of funds into the real economy.  But ordinary depositors ought not be part of this equation.  They are relying on the payment system not the credit allocation system.

Since a well functioning payment system is a public good it ought to be managed separately from the provision of credit.  This would prevent errors in credit risk taking from grinding it to a halt.

If we want to neutralize moral hazard in risk taking, we need to separate the activities so the payment system can happily chug along regardless.  There ought be no risk associated with making a payment system perform.   In other words we need two banking systems not one.  If we want to utilize market “discipline” then make sure it is deployed in a market.  The provision of funds and the taking of credit risk is such a market.  A payment system is a utility not a market.  It doesn’t need discipline.  It needs rigorous operational attention to detail and effectiveness.  It needs administration.

As I said: we need two banking systems.  But I won’t hold my breath