Deutsche Bank: "This Is The $2.5 Trillion Question"

Next Wednesday, the Fed is widely expected to officially launch its balance sheet reduction or "normalization" process, as a result of which it will gradually taper the amount of bonds its reinvests in the process modestly shrinking the Fed's balance sheet.

Very modestly. As shown in the chart below, the Fed's $4.471 trillion balance sheet will shrink by $10 billion per month in October and November, or about 0.4% of its total AUM. Putting this "shrinkage" in context, over the same time period, the Bank of Japan and the ECB will continue adding new liquidity amounting to more than $400 billion. As a result, in Q4 net global liquidity will increase by "only" $355 billion, should Yellen begin "normalizing" in October following a September taper announcement as expected.

That much is known, however there are quite a few unknown aspects about the Fed's upcoming QE unwind, and as a result, Deutsche Bank writes that "the Fed is about to become hugely important for financial assets."

Assuming it all goes well, DB forecasts smooth sailing ahead, manifested by "nominal core rates will be relatively stable and the dollar gently weaker. 10s might trade a sustainably lower range 1.8-2.3 percent. There will be more of a gradual risk asset rotation favoring US (growth) equities, EM, some commodities at the expense of (value) equities, Eurostoxx, NKY with credit somewhere in between."

But, as Deutsche Bank's chief strategist Dominic Konstam writes, "There is a good chance it does not go well."

Here's why. On one hand, Konstam predicts that if the Fed insists on further hikes as it normalizes QE, elevated real rates threaten a more generalized risk off. Even more ominous is that even if the Fed is open to a more relaxed interest rate outlook, there is a looming challenge that balance sheet reduction poses to bank lending. While Fed balance sheet reduction starts small it will be very big come the peak by end 2018 at a maximum pace of $50 billion per month ($30 billion in Treasuries and $20 billion in mortgages). The interest sensitivity of bank demand for securities relative to the non-bank sector "will be absolutely critical as to whether the Fed commits a spectacular own goal on economic growth." For Deutsche Bank, the mere uncertainty around this reinforces a conclusion it made last week when it pointed out that with recession risks the highest in ten years, the Fed should pause tightening; sure enough overnight Konstam doubled down and said that the question markets around the balance sheet reduction "suggests at the very least the Fed should not raise rates in 2018, as the balance sheet reduction ramps."

Or else risk assets - such as Deutsche Bank stock - get it...

As an aside, Konstam touches on something we discussed yesterday, namely the hint by the Bank of Canada that it and other central banks may soon change to their mandate as relates to 2.0% inflation targets. Having demonstrated time and again that they are incapable of hitting their targets, central banks may simply redefine the new inflation target at 1.5% in our lowflation, low r-star world, and declare victory. This to Konstam would be a disaster, especially since officially measured core inflation - when stripping away food and shelter - is now the lowest its has been 2003, and worse: "the dominance of shelter inflation means that much of the inflation that does exist largely reflects a tax on consumers." To wit:

Underlying inflation looks a lot closer to 1 percent or below rather than 2 percent, meaning real rates are already significantly positive when, on their own metric, the short term neutral rate is close to zero. The latest CPI print may give some inflation pessimists reason for pause, but we think it need not. Shelter inflation is running at 3.3% y/y and contributing a whopping 1.4pp to the 1.7% core CPI, while core CPI ex-shelter slid further to 0.5% y/y. Without equity withdrawal by homeowners, the dominance of shelter inflation means that much of the inflation that does exist largely reflects a tax on consumers.

While we may, and almost certainly will have to touch on the inflation issue later, especially if the BOC is indeed set to unveil a change to its policy mandate, a harbinger to abolishing a 2% inflation target across all central banks, something which Bill Dudley has also hinted at last week, for now we focus on the QE unwind.

The problem here, according to Deutsche, is that the Fed has no idea what impact its action will have on the one factor that is most critical for economic growth and contraction (or anything else for that matter): bank lending. To be fair "the Fed understands that there could be a problem." In their Addendum to the Policy Normalization Principles and Plans the Fed states:

“Gradually reducing the Federal Reserve's securities holdings will result in a declining supply of reserve balances. The Committee currently anticipates reducing the quantity of reserve balances, over time, to a level appreciably below that seen in recent years but larger than before the financial crisis; the level will reflect the banking system's demand for reserve balances and the Committee's decisions about how to implement monetary policy most efficiently and effectively in the future. The Committee expects to learn more about the underlying demand for reserves during the process of balance sheet normalization”.

As Konstam tongue-in-cheek writes, "we bet they do and, as innocent a question as it may seem, the underlying demand for reserves is a critical one." He next makes a point we have been pounding the table on since 2010: "Reserves simply are cash for the banks." This is shown in the DB chart below...

... and the Zero Hedge chart which we have shown in some capacity ever since 2011:

Here is DB's explanation why the QE unwind will have a critical importance on commercial bank balance sheets:

When the Fed reduces its balance sheet, someone needs to buy those Treasury and mortgage securities instead. There are the banks, and then there are the nonbanks: foreigners, households, mutual funds, pensions etc. If the banks don’t buy the securities – let’s say they have a fairly low elasticity of demand for more securities, so a small reduction in the price doesn’t lead them to buy more securities – and if the  non-bank sector has a relatively higher elasticity of demand, then deposits in the banking system will fall. Foreigners or funds may switch cash assets for securities that the Treasury is no longer selling to the Fed; the Treasury will trot back to the Fed, hand over the cash, and the Fed will cancel that cash liability to match its reduced asset. The banking system has lost deposits.

This is shown in Figure 7 above, right.

What then do the banks do on the other side?

They could cut their own cash holdings. They could cut their securities or they could cut their loans. They will probably do a bit of everything. The loan deposit ratio would almost certainly rise but given the drop in deposits it is likely that loan growth will be weaker not stronger since the extent to which cash and securities are anyway reduced impairs HQLA. Regulatory relief would help to the extent that HQLA assets may become less binding. Ideally the banks could buy the securities the Fed doesn’t want to own. Then there is a more straightforward switch away from cash to more securities. Deposits would be unaffected. More securities will also boost loan deposits. If deposits are stable this must imply a tendency to raise lending growth. But banks will only buy the securities if they are more interest rate sensitive than the non-bank sector.

Which then brings us to what DB calls the "2 ½ trillion dollar question – the anticipated amount of Federal Reserve balance sheet rundown." Here's why so much depends on how long QE unwind will go for:

[I]ronically less regulatory pressure discourages the banks from buying more securities – that’s a bit of a conundrum. Moreover if there is a low inflation equilibrium and the Fed is raising real rates “too quickly”, who will be more interest rate sensitive? If the dollar is weaker as well, it seems reasonable that foreign investors will eager lap up Treasuries on mild backups – deposits are more likely to be squeezed with loans.

This wouldn’t be such a problem if loan growth wasn't already concerningly weak, something we last highlighted three months ago.

As DB admits, "C&I loan growth looks far closer to what we’d expect in the middle of a recession, having decelerated sharply and momentum only just stabilizing (though not really improving as of yet). If the bank demand proves less rate sensitive than non-bank demand, the drop in deposit could be part of a sharper, more pernicious tightening than intended, with reduced availability of credit and initially higher yields."

In such a world, we expect the initial move would be to higher yields, as bank buying fails to step into the void left by the Fed. This would be to the detriment of the real economy, however, and the medium to longer term implication would be the Fed being forced to recognize its mistake and reverse."

In short: with the most important aspect of any vibrant, healthy economy - namely solid loan growth - lacking, the Fed's reserves, i.e. bank cash, reduction, will have a dire (and direct) impact on futher loan growth, or rather contraction, further exacerbating this metric, and finally pushing commercial bank loan growth into outright contractionary territory, has always been accompanied by a recession.

It almost begs the question: is the Fed now actively seeking to launch the next recession (under president Donald Trump no less)? Perversely, it would make a lot of sense: with the business cycle now broken as a result of so much undue implicit reliance on asset prices, all of which are in a bubble - something which the Fed also understands - it would be beautifully symmetric that the same agent, the US Federal Reserve, that broke the business cycle and unleashed one of the longest, if artificial and on the back of trillions in central bank liquidity, economic expansion is now eager (and hoping) to be the catalyst for the next recession.

Or said simply, is the Fed now eager to accelerate the next economic recession in order to undo its own damage to a business cycle which has forgotten there should also be a contraction?

With all due respect to Konstam and Deutsche Bank, it is this that is the real $2.5 trillion question.