Or so a correspondent tells me. What do we know?
Well, to go insolvent, the Fed would have to recognize losses of apx. $40 billion on its asset portfolio of $4.5 trillion. This is not unthinkable, particularly since it is already holding nearly $2 trillion of risky assets and the several trillion of US securities it is holding are longer term securities. On the other hand, if the Fed is able to hold all of its assets through to maturity, it is not likely to have to recognize losses.
On the other hand, the entire point of conducting monetary policy is that in the event of a tightening, the Fed will have to swap its existing assets with its member institutions (traditional OMO) or with other entities that were not traditionally doing business with the Fed (now, that is worth an entirely new blog post). Also as an aside, much of what the Fed is doing on its liability side is in the reverse repo market (think of these like gigantic collateralized checking accounts that big non-affiliated institutions hold at the Fed). When the Fed tightens, what it essentially does is take securities off the asset side of its balance sheet and swaps them for an equivalent amount of cash or reserve credit from the asset side of a member bank or other institution. From the Fed’s balance sheet perspective, its assets (and hence size of the balance sheet) fall by the amount of the swap, and when it secures the cash or reserve credit, it destroys them, thereby decreasing the monetary base / liability side of its balance sheet.
From the perspective of the banks/institutions being asked to swap reserves/cash for assets, their asset composition changes but there is no net change in their financial position.
HOWEVER, if the value of Fed assets falls and the Fed needs to mop up liquidity/tighten, what happens? Wellllll … then we have a problem. Suppose the Fed has a toxic MBS on its books for $100. It got there by creating deposits/cash from nothing and giving those to other financial institutions in exchange for the MBS. Now, if the value of the MBS falls to, say, $60, if the Fed engages in an asset swap (it may, because these are often collateralized transactions) it would be giving $60 of assets to the banks and making the banks send back $100 in reserves. That’s nice for the Fed … but it causes a $40 capital loss at the banks. In other words, if the Fed has to recognize losses on its assets by engaging in a tightening, it can cause bankruptcies/insolvency in the banking and broader financial/shadow banking system.
How might we deal with that?
(1) Let the banks die? That ain’t going to happen.
(2) Have the taxpayers recapitalize the banks?
(3) Have the Fed give $100 worth of assets in order to mop up liquidity instead of the $60. (think of this in the same way that we saw troubles in the CDS market).
Well both 2 and 3 are going to have similar impacts. If you want the Fed to deliver $100 of value in exchange for $100 of reserve bank credit, it won’t have enough assets on hand to do it if the value of its capital goes negative. So, the question remains, how do you deliver the other $40 that the banks are owed? It can only come from taxpayers, or simply to not do it.
I am leaving out the possibility of an inflation tax for now, and we owe you a discussion of the factors which impact the public’s willingness to hold dollars. But I want to toss out the question for you that my correspondent tossed to me. He claims this is all an illusion because “the Fed can manage its liabilities.” In other words, what is being argued is, to summarize what is otherwise a smoke and mirrors argument, is that the Fed can just take reserves from its member banks and other institutions. So, yes, I agree the Fed can do this, but this is nothing more than an outright tax on the banks, and still leaves us with the situation where Fed solvency issues translate into banking system bankruptcies, and we are back to our original questions.
So, while it is technically the case that the Fed “can’t go bankrupt”, in my view that is true in as meaningless a way as it is possible for it to be true.
Hopefully more on this in the near future.