The Recent Bond Yield And SPX Declines (And Now Bitcoin) Tend To Recur Year After Year

|
Includes: DDM, DIA, DOG, DXD, EEH, EPS, EQL, ETCG, FEX, FWDD, GBTC, HUSV, IVV, IWL, IWM, JHML, JKD, OTPIX, PSQ, QID, QLD, QQEW, QQQ, QQQE, QQXT, RSP, RWM, RYARX, RYRSX, SCAP, SCHX, SDOW, SDS, SFLA, SH, SMLL, SPDN, SPLX, SPUU, SPXE, SPXL, SPXN, SPXS, SPXT, SPXU, SPXV, SPY, SQQQ, SRTY, SSO, SYE, TLT, TNA, TQQQ, TWM, TZA, UDOW, UDPIX, UPRO, URTY, UWM, VFINX, VOO, VTWO, VV
by: Robert P. Balan
Summary

Recent decline in yields was not a random event. It's also seasonality dictated by confluence/divergence of liquidity flows. There's a tendency for yields to fall during late H1 of any year.

Neither was the steep rally in SPX a random event as well. Positive liquidity background amplified the favorable newsflow and will continue to do, pushing equities higher somewhat further.

Barring something dramatic from the Fed or from The Donald, the 2019 yields and equities may rise in next few days (maybe up to two weeks), fall until late August.

Bitcoin assets confirmed the top of May 30/May 31, by falling sharply since then. But a small uptick lasting a few days should take place soon - last chance to make a temporary exit. The next decline seeks a low during Week 2 in July.

There has been multiple explanations and rationalizations of the very recent and sharp decline in bond yields. The most prominently published rationalization was the action taken by billionaire and former hedge fund manager Stan Druckenmiller's piling into US Treasuries as the US trade war with China escalated.

Mr. Druckenmiller was quoted by Bloomberg (here) as saying:

When the Trump tweet went out, I went from 93% invested to net flat, and bought a bunch of Treasuries," Druckenmiller said, referring to the May 5 tweet (the first one) from President Donald Trump threatening an increase in tariffs on China."

He was also quoted as saying:

Not because I'm trying to make money, I just don't want to play in this environment."

He clarified in another interview with CNBC (here), that he kept all his long investments and used other vehicles, including Treasuries, to hedge his market exposure. That made Mr. Druckenmiller go from 93% invested to net flat.

That is the smart money talking. Druckenmiller has a net worth of $5.1 billion. The former chief strategist for George Soros decided to convert his hedge fund into Duquesne Family Office in 2010. His long-term track record, making returns of about 30% a year over three decades, established him as one of the world's top money managers (Bloomberg). So, it is natural to think of the recent sharp fall in bond yields as collective action taken by bond market smarts hedging their risk asset exposures, or betting on falling yields, which equates to rising bond prices. But that would be merely partially correct.

Recent decline in yields and SPX (and now, Bitcoin) were not random events

The fact is, the recent decline in yields was not a random event. It's also a consequence of the seasonality effect dictated by confluence/divergence of liquidity flows. There's a tendency for yields to decline, sometimes sharply, during the late part of H1 of any year, as negative newsflow is amplified by a drought of systemic liquidity starting in early May, in any given year. We discussed this at length in a one-on-one Q&A with Mr. Daniel Shvartsman, Director of Seeking Alpha's Marketplace, which was published on May 25, 2019 ("Looking At Systemic Liquidity"). As a way of introducing the topic, we said:

One example I would like to show is how the seasonality of the values of the Treasury Cash Balances closely resemble the seasonality of changes that you see in bond yields over a historical period (say, a five-year average). That holds as well over any particular period, for instance, like what we have seen so far in 2019. The bond yield performance this year, so far, has not diverged much from the five-year average of the 10yr yield. Or for that matter, from the influence of what we call the "Treasury Model". (The chart below illustrates the symmetric tendency of moves in TCB and bond yields).

The consistency of the covariance between the 5-year average of the 10yr yield and the Treasury Model is due to the persistent seasonality characteristics of the Treasury Cash Balance (TCB) in its raw, seasonally unadjusted, form. The Treasury does not vary the schedule of the expansion and shrinkage of the TCB, although the amounts involved in any year could vary widely (see chart above).

Bond Yields

The US Treasury maintains a cash balance to buffer unanticipated and short-run differences between receipts and expenditures at the Federal Reserve. This net amount is what has been called the "Treasury Cash Balance." Treasury balances exhibit significant trends, building up when receipts exceed disbursements and running down when disbursements exceed receipts. As the chart above illustrates, there is also a very distinct seasonality in the movement of the TCB net amount, which has great influence on bond yields, historically.

For instance, the normalized 10yr bond yield profile this year will probably be not much different from its average profile of the past five years. In fact, what is happening at this juncture in yields had happened in 2016 as well. There is a tendency at this time of the year for 10Yr yields to fall sharply, recover, then fall again until late August, with an interim trough in early July (see chart above and below). That recurring pattern is what leads us to say that the current decline in yields was not a random event. It is a consequence of a seasonal trend which is dictated by confluence and divergence of liquidity flows at this time of the year. The dearth of liquidity at this time amplifies any negative newsflow that comes along, causing sharp downside moves given appropriate nudge.

To show the relationships, let's clean up the yield chart, leaving just the primary movers and show the data in terms of five-year averages. No bells and whistles - just the current asset price, the five averages that matter, and the current status of the flows that are being front-run by institutional traders.

Bottom line: So barring something dramatic from the Fed or from The Donald, the 2019 yields should rise up over the next few days (can be up to two weeks), fall again until late August (with interim trough in early July), and then sashay towards a high going into the year-end (see chart above).

This pattern also tallies well with our outlook of the US GDP growth falling until September, then rising to an inflection point in Q1 2010. (See that here and here).

The Treasury Cash Balance is merely one of several major sources of systemic liquidity operative in the financial markets. The chart below shows some of the liquidity factors that matter to equities and for other risk assets, including what we call the "SOMA Model".

Equities

The SOMA model in the chart above encompasses the Open Market Operations being carried out by the New York Federal Reserve in managing and administrating the Monetary Base (MB). The MB, of course, is primarily composed of the Fed's balance sheet and other components (currency in circulation, currency in bank vaults) which amount to small fraction when compared to the size of the assets acquired via the Large-Scale Asset Purchases (LSAP) during the Quantitative Easing during the Great Financial Crisis. The MB also includes the commercial banks' reserves, held at Federal Reserve. The bank reserves are on the liability side of the ledger.

The SOMA model tends to be more suited for projecting future equity prices. The chart below is a treatment of the SPX, which we are showing in nominal and change-rate form. The change rate is essential in understanding the impact of the liquidity flows (which are in change rates). Otherwise, you will need a very strong imagination or very high pattern analysis acuity to see the relationships.

The black, dotted line is the 5-yr average of SPX. The time series is adjusted to remove the front-run effect, and you can see how this SPX 5yr average in juxtaposition with the five-year averages of the SOMA and Treasury Models (brown and green lines, respectively) synched with each other very well.

The key for this kind of analysis to work is to normalize the time series according to a number of optimized means. That way, it is easy to see that there is almost nothing random about the SPX trends (except of course in the very short term) and that the market tends to do almost the same sequences of moves every year, but in varying levels of equilibrium. That conclusion seems to be supported by the 5yr average profile of SPX hewing very close to the 5yr average of the SOMA model, and for that matter, to the 5yr average of the Treasury Model.

Bottom line: Same as in yields - equities should rise up over the next few days (can be up to two weeks), fall again until late August (with an interim trough in early July), and then sashay towards a high going into the year-end.

Bitcoin

Bitcoin assets confirmed the top of May 30/May 31, as they have been falling sharply since then. But a small uptick lasting a few days should take place very soon, a last chance to exit long positions. This uptick follows an upside blip in liquidity which is also temporarily pushing up equity prices, but it is clearly counter-trend for equities and has the same (but lagged) implication for Bitcoin assets. The next decline in equities and Bitcoin should be significant and could last until Week 2 in July 2019.

Zoomed in view:

Bottom line: Note that the PAM Bitcoin/Ethereum Model (red line) has already made an inflection point higher, following the lead of the PAM liquidity models. Our interpretation is for a sharp decline (as indicated by the liquidity models) very soon, but it should be brief. The inflection point following the next decline should come during the second week of July. The PAM Bitcoin/Ethereum model leads the BC assets by a few weeks.

The next rally in Bitcoin assets from the July trough should be significant and could make new highs for the year.

A more detailed treatment of these topics was first discussed and provided to the Predictive Analytic Models (PAM) community, on January 5, 2019. You may see this report and other prior but similar articles at Robert P. Balan's blog (here)

Disclosure: I am/we are long TQQQ, USLV, UUP, USO. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.