During January and February, spot gold rose 8.34% (from $1,152.27 to $1,248.33) and spot silver increased 15.08% (from $15.92 to $18.32). During early March, however, precious metals suddenly reversed course, with spot gold declining 2.30% and spot silver declining 5.29% through respective March 15 closes of $1,219.68 and $17.35. Without question, the greatest variable affecting precious-metal performance during recent weeks has been market handicapping of the Fed’s March 15 FOMC meeting. On 2/22/17, Bloomberg consensus expectations for a rate hike at the March meeting measured 34%. Ten trading days later (3/8/17), this percentage had swelled to 100%.
We attribute this swift shift largely to a short stretch of particularly impassioned Fed jawboning, book-ended by the FOMC’s two crucial thought-leaders, Vice Chairman William Dudley and Chair Janet Yellen. On 2/28/17, Mr. Dudley commented, “I think the case for monetary policy tightening has become a lot more compelling,” but then raised eyebrows with uncharacteristic frankness about U.S. asset prices: “There’s no question that animal spirits have been unleashed a bit, post the election. The stock market is up a lot.” By 3/3/17, Chair Yellen sealed the deal for a 3/15 hike in a speech to the Executives’ Club of Chicago, in which she remarked, “Indeed, at our meeting later this month…a further adjustment of the federal funds rate would likely be appropriate.” As avid students of Fed communication, we find the Fed’s tone change since 2/28/17 nothing less than abrupt. What factors account for this sudden shift to urgency following years of trademark caution? Might the Fed be reacting to strengthening economic data? While “soft” economic data and sentiment measures have broadly improved since Trump’s election, “hard” economic statistics (historically favored by the Fed) have remained stubbornly weak. During the recent period of hawkish Fed rhetoric (2/28/17-3/3/17), soft data continued to surprise to the upside: the Chicago PMI registered 57.4 versus estimates for 53.5 (2/28); Conference Board consumer confidence came in at 114.8 versus estimates for 111.0 (2/28); and ISM indices for manufacturing (3/1) and services (3/3) totaled 57.7 and 57.6 respectively (versus estimates for 56.2 and 56.5). Hard data released during the same period, however, continued to disappoint: Q4 GDP (2/28) was adjusted downward to 1.9% versus estimates for 2.1%; wholesale inventories (2/28) fell 0.1% versus estimates for an increase of 0.4% (pressures Q1 GDP); the U.S. trade deficit (2/28) surged to the second worst reading since 2008 (expanding to $69.2 billion versus estimates for $66.0 billion); and construction spending (3/1) declined 1.0% versus estimates for a gain of 0.6%. On 3/15, the Atlanta Fed’s GDP Now forecast for Q1 2017 collapsed all the way to 0.9%, after registering 2.5% as recently as 2/27 and 3.4% on 2/1. As shown in Figure 1, the spread between hard and soft economic data surprises has now expanded to a 17-year high (Bloomberg). Something has to give, and history overwhelmingly suggests soft data and sentiment measures are likely to recede dramatically in coming weeks. The Fed is well aware of these probabilities, yet still felt heightened urgency for a March hike.
Our explanation for newfound Fed resolve is that we believe the Fed is finally becoming exorcised over the accelerating bubble in U.S. financial asset prices. This past week (3/9/17), the Fed published its Q4 2016 Z.1 Report (Financial Accounts of the United States). According to the Fed, U.S. household net worth increased a staggering $2.043 trillion during the final three months of the year. By way of context, during the same period, the U.S. Bureau of Economic Analysis informs us that U.S. nominal GDP expanded by some $180.20 billion. During Q4 2016, therefore, U.S. household net worth increased at a rate 11.33 times the rate of U.S. GDP growth.
This ratio is patently absurd and seems to have commanded the Fed’s immediate attention. Unfortunately, the Fed’s apparent alarm over untethered U.S. financial asset prices is coming far too late to absolve FOMC participants of responsibility for yet another Fed induced boom-bust cycle. After all, the gaping detachment of U.S. household net worth from underlying U.S. productive output has been in full swing for almost eight years. Since Q1 2009, the Fed informs us that U.S. household net worth has increased (through 2016) an astonishing $38.016 trillion (from $54.790 trillion to $92.805 trillion) during a period of time in which the BEA calculates a $4.766 trillion increase in nominal GDP (from $14.090 trillion to $18.856 trillion). During the past seven-and-three-quarter years, therefore, U.S. household net worth has exploded at a rate exactly eight times the rate of underlying GDP growth.
As the S&P 500 Index sets new all-time highs on a weekly basis, investors may view discussion of HHNW/GDP ratios as little more than academic flagellation. We beg to differ. There are many relationships of which we are unsure, but we are quite certain no society can increase wealth eight times faster than output forever. Indeed, we attribute gold’s ascent during the past 16 years to progressive recognition, at the margin, of precisely these imbalances. The Fed’s Q4 2016 Z.1 Report suggests U.S. household net worth ($92.805 trillion) has now reached an unprecedented 492% of GDP. To lend historical context, this ratio now rests 40% higher than the 353% average during the five decades prior to the Greenspan/Bernanke/Yellen era (Figure 2). Because the Fed has relied on growth in HHNW as a potent transfer mechanism for “stimulative” QE and ZIRP policies, we would interpret current resolve to hike rates amid faltering economic growth as clear signal the Fed views reigning financial asset prices as increasingly problematic. Because the Fed has a highly checkered history in evaluating negative impacts of pierced bubbles (generally of their own creation), we offer some perspective on how precarious current U.S. asset valuations may be, and how far they may need to fall to rebalance the U.S. financial system toward historical norms.
It stands to reason that in a normally functioning economy, household net worth should expand by some factor of GDP growth-plus-savings. The greater the savings rate, the greater the rate of capital formation, leading to higher household net worth. Our examination of history and a dose of logic suggest to us that a reasonable approximation of this relationship might be (Real GDP Growth Rate + Net National Savings as % of GDP) / Real GDP Growth Rate = Projected HHNW/GDP Ratio. In Figure 3, we apply our representative formula to actual GDP and savings growth-rates in decades since 1950 to calculate a reasonable expected rate of wealth formation.
During the 1950’s and 1960’s, as real GDP averaged 4% growth and net savings grew by over 10% of GDP on an annual basis, the U.S. was experiencing what we would view as high-quality capital formation. Our formula suggests any society which can build wealth at such prodigious rates should enjoy a net worth-to-GDP ratio approaching four, which was roughly the case in the U.S. throughout these decades, as reflected in the Fed’s reported ratios in Figure 2. In subsequent decades, our formula continues to generate ballpark-type approximations of the Fed’s reported HHNW-to-GDP ratios, all the way until the 1980’s, when declining growth-rates