Gold Flat After Wild Ride

August proved difficult for the precious metals complex, although gold bullion¹ ended the month flat; the yellow metal remains underwater at -4.46% YTD through August 31, 2021. Gold mining equities³ were hit hard in August, losing 6.05%, and are down 9.53% YTD as of August 31. Silver² lost 6.26% in August and is down 9.50% YTD. 

Month of August 2021

*Mo % & YTD % Chg are expressed in basis points (BPS) rather than percentages. 

Spot gold was essentially flat for the month, falling $0.57 (-0.03%) to close at $1,813.62 for August after a wild ride. On August 6, the NFP (Non-Farm Payrolls) had a stronger than expected number of 943,000 versus the 877,000 consensus estimate. Gold sold off $41.38 on the NFP print, a surprisingly sharp move considering that employment numbers are lagging indicators and generally volatile. Given that more than 9 million jobs were lost during the COVID pandemic, these new jobs are "recovered" rather than "created," a qualifying fact well communicated by the Federal Reserve (Fed).

The early-August sharp drawdown in gold bullion was attributed to the fear of the Fed's tapering timeline being moved forward. However, we did not see a commensurate tapering-fear response in the bond, stock or credit markets to support that view. Belatedly, the Friday, August 6 sell-off was tame compared to what happened early Sunday evening on August 8. Before 19:00 EST, an estimated $4 billion in gold contracts were unloaded into one of the lowest liquidity periods possible. The effect was a flash crash in gold as bullion fell $73.26 from the Friday close, or $114.64 from the Thursday close at the lowest point.

The question remains, why would someone try and sell about $4 billion of gold contracts on an early Sunday evening in August, knowing that liquidity would be near non-existent? We heard several reasons (fat finger, forced liquidation, short selling), but none made much sense in the absence of all the details. Typically, a flash crash of this type is not sustainable once liquidity returns, during regular trading hours, without a solid reason to sell down further. By the end of the week, gold had recovered from the flash crash. The higher weekly close (after the flash crash) also triggered several technical bullish reversal patterns, which caused some buy programs to kick in and some short positions to be covered. After the Fed's Jackson Hole dovish taper message (Jackson Hole Economic Symposium, August 27-28, "Macroeconomic Policy in an Uneven Economy"), gold recovered the NFP print-induced sell-off and flash crash, capping off a wild August (see Figure 1).

Figure 1. Gold Bullion: Flash Crash and Recovery Back Above Trendline and 200-Daily Moving Average 

Source: Bloomberg. Data as of 8/31/2021. For illustrative purposes only.

Miners Were Hit Harder than Bullion in August

However, the sudden drawdown in gold impacted gold stocks and silver (see Figure 2) through triggered stops and further liquidity reductions. Both silver and gold mining equities have drifted following the June Federal Open Market Committee (FOMC), but the flash crash saw liquidity fall away further. Liquidity had been in decline prior, and mysterious Sunday night flash crashes will only increase investor wariness. The pending Fed announcement on the taper schedule was also not helpful. The tapering concern had weighed disproportionately on the precious metals complex due to the memory of the 2013 Taper Tantrum. Last month, we highlighted the reasons why this coming taper is nowhere close to the 2013 experience. Chair Jerome Powell's August 27 Jackson Hole speech further cemented the view that this current tapering and tightening phase will be much different and more benign than the 2013 version.

In August, stocks lost market interest and smaller gold mining companies underperformed larger-cap stocks for most of the month. The sector was whipped around by the month’s gold price action, even despite the improved financial conditions mining companies demonstrated during their Q2 2021 profit reports. Many companies are paying dividends with stock yields of over 2% and numerous firms have more cash than debt on their balance sheets. Markets may fear declining margins and a diminished profit outlook for precious metals stocks, which has not been the case thus far, although cost pressures are rising.

We continue to believe that the sector’s valuations should start to compel merger activity. Furthermore, observing investors are likely to re-rate gold mining stocks more appropriately, given their attractive financial profiles and the profits that the gold price can generate. At current valuations and relative to the risky valuations of financial assets in general, we believe that continued exposure to gold mining stocks should prove quite rewarding.  

Figure 2. Silver and Gold Mining Equities Retesting Major Support Levels (2018-2021)

Source: Bloomberg. Data as of 8/31/2021. For illustrative purposes only.

A Kinder, Gentler Taper

The markets interpreted the Fed's message at the Jackson Hole Economic Symposium as very dovish, and stocks made new highs, yields fell, the U.S. dollar (USD) weakened, and gold rose over $25 on the day. Powell essentially repeated the dovish July FOMC minutes and avoided making any waves as the market was hyper-focused on the Jackson Hole meeting. The short message was that it could be appropriate for the Fed to begin tapering this year and that the timing of tapering would not be related to the timing of the first rate hike. Jackson Hole did not point to any signs of a September announcement, making November or December the most likely taper announcement date. Between now and then, there will be two additional jobs reports and new economic data on the impact of the COVID delta variant that will likely influence the Fed's taper announcement.

The Fed continued to view the current inflation pressures as being driven by COVID-related supply and supply-chain issues which are, in their view, still transitory. There was no hint of the 2% AIT (average inflation targeting) goal changing. The aim remains to run the economy hot to achieve full employment and improve the outlook for disadvantaged social groups. Since the Fed still has not made "significant further progress" in this area, any future taper will be gentle, and the Fed is likely to remain highly accommodative for some time after that. With no immediate Fed action on inflation, nominal rates will not likely spike (at least due to the Fed), and inflation expectations may grind higher. The result is that negative real interest rates will stay lower for longer. In turn, this will keep a cap on the USD. The long-term dual themes of financial repression and dollar debasement continue to be reinforced with every significant Fed speech.

Gold Hurt by Tapering/Tightening Question

Gold has been constantly buffeted by the taper/tightening fears throughout the year, more so than most asset classes. In the long run, the potential impact on gold from fewer billion dollars of asset purchases per month seems way overblown when considering the scale, scope and future direction of central bank asset purchases. Figure 3 highlights the major central bank balance sheets as a percentage of GDP (gross domestic product).

How many believe these numbers have any hope of materially decreasing in the next several years? As a reminder, the Bank of Japan (BOJ) went to a ZIRP (zero interest rate policy) stance over 20 years ago, and the European Central Bank (ECB) went ZIRP nearly ten years ago. The BOJ and ECB never left the ZIRP world at any point; they went deeper into NIRP (negative interest rate policy). The ECB went to a NIRP in June 2014, and the BOJ went to a NIRP in January 2016. The Fed is just barely one year into ZIRP.

Figure 3. Major Central Bank Balance Sheets as a % of GDP (2008-2021)

Source: Bloomberg. Data as of 8/31/2021. For illustrative purposes only.

Negative Yielding Bonds Still Expanding

There is $15.7 trillion worth of global negative-yielding bonds despite more than $30 trillion worth of global fiscal and monetary stimulus since the GFC (Global Financial Crisis). Global quantitative easing (QE) prevents yields from rising, keeps the amount of negative-yielding bonds increasing, and forces deeper negative real interest rates. Global central banks will need to keep expanding balance sheets to support their QE programs and maintain low yields; this will provide the global economy with continuous accommodative conditions that support growth.

If inflation shows up, the Fed will have two choices. The first would be to tighten to slow the recovery, but this seems less likely after the Jackson Hole meeting. The second would be to risk real inflation and allow long-term inflation to become entrenched in consumer mindsets. The problem is that virtually every major asset class is priced off of the ultra-accommodative policy. If the Fed is forced to address real inflation by raising interest rates, its actions will be a highly destabilizing risk factor. Longer and lower negative real yields are the path of least resistance. Globally, negative-yielding bonds will continue to expand. Gold will provide optionality in an inflation overshoot and also in a Fed-driven risk-off event.

Gold Will Brighten as Real Yields Grind Lower

In Figure 4, we overlay the total amount of negative-yielding bonds with an estimate of total gold holdings (includes ETFs, CFTC⁷ and managed futures). Since January 2016, the annualized growth rate of negative-yielding bonds is about 20% annualized. The growth in total gold holdings is a similar 20% annualized. The R-square⁸ since January 2016 is over 76%. Negative yielding debt is now about 23% of total global debt outstanding, up from 8.6% from the start of January 2016. We continue to see gold as a core holding growing in the medium- to long-term as the amount of negative-yielding bonds increases and real negative yields grind lower.

Figure 4. Negative Yielding Bonds and Gold Holdings (2015-2021)

Source: Bloomberg. Data as of 8/31/2021. For illustrative purposes only.

Cutting Through the Noise with Gold Models

Over the decades since the end of the Bretton Woods system of monetary management,⁹ the highest correlating variables to gold bullion have been M2¹⁰ money supply, the USD and 10-year real yields. Using monthly data, running a multiple regression analysis of these variables versus gold produces an R-squared⁹ of 88% over the past 20 years. One can then create a "Basic Gold Model," as shown in Figure 5. The Basic Gold Model continues to trend higher because M2 is making new highs. M2 is growing at a 14% annualized rate even after the COVID emergency spending, which is twice the 50-year average growth rate. The 10-year TIPs (Treasury Inflation Protected Securities) yield has made recent all-time lows, and U.S. Dollar Index (DXY) remains at the low end of the weekly price range. The bullish takeaway would be that gold will eventually move up towards the historical price relationship. The bearish counterpoint would be that this historical relationship may no longer be valid, and other variables are now more critical (we do not see evidence of this yet).

The Basic Gold Model in Figure 5 is missing inputs for positioning flows and "market conditions" (a catch-all phrase for the multitude of factors affecting gold relative market attractiveness). Though M2, DXY and TIPs yield have some embedded or implied "market conditions effects," it is insufficient to capture all the nuance of price actions. If we add some market and positioning variables/data, we can create a better model. This "Enhanced Gold Model" is a better model with 90% R-square, and visually it is a better fitting model. It also indicates that market conditions and positioning are currently outweighing the effects of M2, DXY and TIPs yield in the short term. Market conditions and positioning are intertwined to some degree but are short-term in nature and to use a current overused word, transitory. The lower panel is the Basic Model to the Enhanced Model ratio. When this ratio is elevated as it is currently, it is an indication that market conditions and positioning are likely discounting gold too bearishly (or too extremely) and are at risk for a mean reversion move. This ratio is not a timing model, but it is an indication that the pendulum has swung too far.

Figure 5. Gold Models: Long-Term Drivers and Market Condition/Positioning

Source: Bloomberg. Data as of 8/31/2021. For illustrative purposes only.

All Good Things

Growth indicators are slowing, consumer confidence has experienced a sudden drop in the U.S. and the international slowdown is accelerating (especially in China). As the surge in monetary and fiscal stimulus effects peaks and crests, in the not too distant future, there will undoubtedly be a growth scare that may delay or defer tapering and rate hike expectations and even possibly add new stimulus measures. If this were to occur while inflation was still reasonably hot, fears of stagflation would start to percolate. As we edge closer to tapering, the next question will coalesce around whether the Fed can start to tighten without severe market disruptions. The Fed's recent 2018 experience and the medium- and long-term experience of the ECB and BOJ all point to a firm no.