WILL MARKETS HAVE A LIQUIDITY CRUNCH? / THE FINANCIAL TIMES

Will markets have a liquidity crunch?

Only if the Fed doesn’t blink first

Robert Armstrong

© FT montage/Reuters

Liquidity worries

Two weeks ago, while I was away, the wise Kate Duguid wrote in this space that people were worried about imminent declines in financial system liquidity. 

The worries grow from two important changes.

Change one: Until last spring, the Federal Reserve was buying bonds, which pushes cash into the financial system. 

Now it is selling bonds (on a net basis), which sucks cash out of the financial system at a rate of about $95bn a month.

Change two: in the run-up to the debt ceiling deadline, the US Treasury could not issue new debt. 

So it funded its spending by simply running down its checking account (known as the Treasury General Account, or TGA). 

That releases liquidity, too. 

When the TGA runs down, the government is putting cash into the economy by spending, but not pulling it out by issuing new debt or raising taxes. 

Now the TGA, standing at just $50bn or so, needs to be rebuilt. 

Yesterday, the Treasury said it was aiming for a TGA balance of $425bn by the end of June and $600bn by the end of September. 

That’s the government taking cash out of the system by selling debt and not replacing it with spending; a cash drain.

As Kate pointed out, there are two related worries about the liquidity decline. 

The first is that it will lower the level of bank reserves, leaving banks less able to deal with market turbulence, just a few weeks after a close scrape with a banking crisis. 

The second worry is that withdrawing liquidity from the financial system makes risky asset prices fall.

The worry about banks is a little hard to understand, because in one sense, when Treasuries are bought and sold, banks are nothing more than conduits. 

Say I’m an asset manager and buy $1mn in newly issued Treasuries through my bank, which (let us assume for simplicity) is a Fed primary dealer. 

The bank swaps $1mn in its reserves for the Treasuries, which I now own. 

But that $1mn decline in the bank’s reserve assets is matched with a $1mn reduction in its liabilities — I used a deposit at the bank to pay for the Treasuries. 

Bank equity, both at the individual bank and the system levels, is unchanged.

What’s the worry, then? 

As I understand it, because banks have much fewer liquid assets (cash, reserves) than deposits and other liabilities, reducing both liquid assets and liabilities by the same absolute amount leaves liquid assets in the system at a lower level in proportion to liabilities. 

The system is therefore less able to withstand more withdrawals. 

In addition, liquidity is not evenly distributed around the system. 

If one bank sees a lot of customers swapping deposits for Treasuries it’s going to need more liquidity. 

The point is that, one way or anther, new bank liquidity is going to have to be raised, at a cost.

How worried should we be about the banking system as liquidity falls? 

Not very. 

The pseudo-crisis regional banks just went through was not a systemic liquidity problem at all; it was a balance sheet mismanagement problem at a handful of very badly-run banks. 

And as the irreplaceable Joseph Wang (aka the Fed Guy) pointed out to me, while that problem did cause a wave of withdrawals at other banks, those other banks were able to source new liquidity, from the Federal home loan banks or elsewhere, without much trouble. 

In other words, the system just had a liquidity stress test of sorts, and it passed.

That leaves the asset price worry, which is very much alive in the minds of Wall Street analysts. 

This from a story that ran in Bloomberg over the weekend:

JPMorgan Chase & Co. strategist Nikolaos Panigirtzoglou estimates a flood of Treasuries will compound the effect of QT on stocks and bonds, knocking almost 5% off their combined performance this year. 

Citigroup Inc. macro strategists offer a similar calculus, showing a median drop of 5.4% in the S&P 500 over two months could follow a liquidity drawdown of such magnitude, and a 37 basis-point jolt for high-yield credit spreads.

I think the idea that pulling cash out of the financial market drives asset prices down makes a lot of sense. 

Go back to the asset manager example: where once I had $1mn in cash, I have have $1mn in Treasury bills yielding five per cent. 

All else equal, I am now going to be less inclined to look for risk assets to buy, because now I have some yield and some rate risk.

Or you can think of it in a simpler way, which Steve Blitz of TS Lombard put to me. 

When the Fed is out there buying Treasuries, someone else in the private sector is not buying them. 

When the Fed is not buying Treasuries, and the Treasury is also selling more of them than usual to refill the TGA, that private someone is buying them, presumably in lieu of some other asset. 

And “everyone takes a step to the left on the risk asset scale”, depriving riskier assets of buyers and driving prices down.

The problem is quantifying this and assessing the timing. 

I have no idea how you turn the general idea that liquidity and risk appetites are related into a forecast. 

So I’m comfortable saying the imminent liquidity decline is a headwind for risk asset prices, but I am not comfortable being much more precise than that (if anyone has a good quantitative model, by all means send it along.)

Is a liquidity-driven decline in asset prices tradeable? 

Courtney Rosenberger and Chris McGrath of Strategas think so. 

They take the view that liquidity pushed investors out the risk curve a step further, arguing that when liquidity is rising growth stocks outperform (as they have as the TGA was being run down this spring) and value stocks outperform when liquidity is falling. 

They plot their net liquidity indicator (which combines the liquidity effects of the Fed’s Balance sheet, its reverse repo facility, and the TGA) against the relative performance of growth and value indices:

We’ll see if that pattern holds between now and September. 

But there is another question to be answered first: will the Treasury and the Fed allow liquidity to fall if market stress increases? 

As Blitz of TS Lombard pointed out to me, cash that is absorbed by the sale of Treasuries or the by building up the TGA is not destroyed. 

It is stored, and can be released again by putting QT on hold or building up the TGA more slowly, should circumstances warrant caution.

Michael Howell of CrossBorder Capital suspects that this is exactly what is going to happen. 

Building up the TGA to $700bn “is on the wish list, but can they do it? 

I think not,” he says. 

That the Fed is seriously concerned about having adequate liquidity in the system is shown by the fact that it is raising cash now by selling short-term bills, rather than longer-dated Treasuries. 

The idea is to entice money market funds, which now have cash parked in the Fed’s reverse repo programme, to buy bills instead. 

If this happens, banks will be intermediaries in less Treasury purchases and bank reserves will decline less. 

Is the Fed really going to commit an “own goal”, in Howell’s words, by draining liquidity from a banking system it has gone to some length to protect?

The question for investors, then, is exactly how much market stress the Fed and the Treasury are willing to tolerate before they put the cork back into the liquidity drain. 

The answer might determine which asset classes outperform for the rest of the year.