The popularity of gold jewellery in China and India has plummeted in recent years, dragging global demand to its lowest level in seven years. But despite this massive decline in the largest single source of gold demand, prices have proved resilient, largely due to strong investor demand. In the face of ongoing interest-rate hikes by the US Federal Reserve, ETF holdings have risen to levels last seen in 2013, a period defined by much higher gold prices.
Prevailing prices of nearly $1,300 an ounce have limited fundamental support. Continued rate hikes threaten to reverse last year's near-record ETF inflows, as still-manageable inflation would push real interest rates well above levels that attract investor interest.
Morningstar equity analysts expect gold prices to fall in the remainder of 2017 and into 2018, before increased Chinese and Indian jewellery demand catalyses a rebound. Investor outflows can strike suddenly, but a full recovery in jewellery sales will take time.
As they have a habit of doing, gold markets have surprised many analysts over the past couple of years, including us. In particular, the sources of demand strength and weakness have upended many forecasts - investor demand proved resilient in the face of gradually rising interest rates, and once surging emerging-market jewellery demand crested and fell.
ETFs stocked up on gold in a big way in 2016. It was the second-largest accumulation of gold by ETFs in a calendar year, lagging only 2009, which was surprising, since rising real interest rates have historically been associated with ETF outflows.
Yet we doubt the relationship between real rates and investor gold demand is irreparably broken. Rising real interest rates will increase the opportunity cost of owning gold, which should eventually trigger ETF outflows and pressure gold prices. We continue to view ETF outflows as the biggest near-term threat to gold prices. For example, in 2013, as investors liquidated nearly 25% of their holdings in gold ETFs, gold prices fell by roughly $300 an ounce, or about 25%. Gold miner stocks followed suit, with most names plunging 50% to 70%.
Strong Fundamentals Drive Gold Price
While investor demand has strengthened over the last two years, jewellery purchases have slowed significantly. In 2016, global gold jewellery purchases totalled just 1,891 metric tons, down more than 23% from 2014. Despite weakness in the two largest gold jewellery markets, China and India, we think underlying fundamentals remain strong. As governments pull back on policies that hamstrung gold buying and as household incomes continue to rise, we expect China and India to drive long-term growth in gold jewellery demand.
Investment demand and jewellery have had the largest impact on the current balance of gold supply and demand. We expect recent trends - rising investor demand and falling jewellery demand - to reverse in the years to come.
The timing of these two reversals portends near-term pressure on gold prices. We expect investor demand to weaken amid continued rate hikes, which should result in a significant outflow of gold ounces from ETFs. Not only does lower investor demand result in weaker demand for gold, it also results in excess supply, as these ounces are sold back into the market. As a result we forecast gold prices will be weak in the near term, declining to $1,150 an ounce by the end of 2017.
In the longer term, provided real rates remain above zero, investor demand is unlikely to return in a big way. However, the foundation for growing jewellery demand is still there. We expect growing demand from China and India will drive stronger overall gold demand in the future.
In recent years, mine production has continued to grow. Amid efficiency drives, lower oil prices, and weaker producer currencies, mine costs have fallen over the last three years, allowing higher-cost mines to remain in production. Falling production costs have helped support production growth, despite weaker gold prices.
However, through the first half of 2017, oil and currency trends appear to be reversing and may push production costs higher, constraining supply and lifting prices.
The precipitous decline in oil prices continued into 2016, with Brent crude oil prices down roughly 45% from 2014 levels. However, oil has since recovered somewhat, with the latest spot prices for Brent up 11% from the 2016 average. Morningstar’s energy team forecasts Brent prices to reach $60 a barrel by 2020, 35% higher than today’s spot price. All else equal, higher oil prices should push gold production costs higher.
Strong Dollar Impact
Currency movements also helped reduce gold production costs. In 2016, key producer currencies like the Canadian dollar, Australian dollar, Chinese renminbi, and Russian ruble weakened by 3%, 1%, 4%, and 9% against the U.S. dollar, respectively, extending multiyear trends. Weaker local currencies effectively lower US-dollar-denominated mining costs. However, this deflationary effect on costs appears to be fading. Through the end of July 2017, the Canadian dollar, Australian dollar, and Russian ruble have actually strengthened by 7%, 10%, and 3% against the US dollar, respectively.
Despite recent years of improved free cash flow across gold companies, we note that this hasn't yet resulted in significant new-mine investment. Meanwhile, increased production has depleted the industry's reserves. Reserves have fallen by 13% for the gold miners we cover, leading to a 6% decline in the implied reserve life. Though exploration could increase reserves, investing in new mines will be required to boost supply. We continue to expect undersupply to incentivise the development of new projects, and we reiterate our long-term gold price of $1,300 by 2020.
Historically, a positive real interest rate has been negative for gold prices, as it raises the opportunity cost of holding the yellow metal. This can be seen in the historical relationship between real interest rates and the gold price. Only when real interest rates dipped into negative territory in 2011 did gold prices approach nosebleed heights. Not long after, real rates returned to positive territory in 2013, and gold prices fell dramatically.
With the federal-funds target range now at 100 to 125 basis points, the real cost of holding gold is higher than prior years, when the Fed-funds rate was near zero. While the real interest rate has flirted with negative territory, it's remained at about 40 to 50 basis points since July 2016.
Despite what appears to be a weakening investment case, ETF holdings remain just as high as they were before the Federal Reserve set out on its path of rate hikes. ETF holdings stand at levels seen amid the much higher gold prices of early 2013. Those holdings seem likely to be released back into the market if the Fed maintains its course of rate hikes.
There's ample precedent for a sudden surge in ETF selling brought about by rising rates. During 2013, ETFs experienced massive outflows of gold, nearly 750 metric tons, or a 36% decline from the start of the year. This was primarily driven by rising yields on 10-year US Treasuries, which increased the opportunity cost of holding gold; 10-year Treasury yields rose more than 100 basis points in just a few months.
Yet while the real interest rate remains at the levels that drove outflows in 2013, gold ETF holdings remain elevated. Geopolitical uncertainty continues to be a significant driver: Bridgewater Associates founder Ray Dalio, for example, recommends that investors put 5% to 10% of their portfolios in gold because of this.
Geopolitical uncertainty is a material risk that is often quite difficult to measure. Gold can provide some safe-haven potential in volatile periods, particularly since it's less exposed to any single country compared with alternatives like currencies or government securities. But the appeal of gold as a safe haven depends on a bit of circular logic: It only provides a safe haven if enough investors believe it does.
Given that we can't predict the timing, severity, or likelihood of major geopolitical events, we do not explicitly use it to shape our forecasts. More importantly for long-term investors, while geopolitical risks can provide a short-term lift to gold prices, they provide little long-term support. Even the impacts of the biggest events fade with time, while the banality of supply and demand endures.
Fed Will Continue to Hike Rates
The Federal Reserve’s statutory mandate focuses on two key metrics, maximum employment and stable prices, and we don’t see any developments that should stall its plan for continued rate hikes.
The Fed’s estimate of a long-term natural rate of unemployment is about 4.7%, a target that the central bank is comfortable with and that represents the necessary slack for the labour systems to function appropriately. When unemployment is meaningfully higher than 4.7%, we would expect the Fed to employ more accommodative policies. As of July 2017, the Bureau of Labor Statistics estimates that unemployment in the US is just 4.3%.
In addition, the latest commentary from the Federal Reserve's Federal Open Market Committee (FOMC) states that the bank expects labour market conditions will tighten even further. "Information received since we met in May indicates that the labour market has continued to strengthen and that economic activity has been rising moderately so far this year," the FOMC wrote in a June 2017 statement. "Job gains have moderated but have been solid, on average, since the beginning of the year, and the unemployment rate has declined."
The Inflation Factor
With employment performing better than targeted, inflation remains the bigger concern for the FOMC. In its June statement, the committee noted that inflation had declined and was running somewhat below the long-term target of 2%: “On a 12-month basis, inflation has declined recently and, like the measure excluding food and energy prices, is running somewhat below 2%. Market-based measures of inflation compensation remain low; survey-based measures of longer-term inflation expectations are little changed, on balance.”
Inflation, as measured by the Bureau of Economic Analysis’ personal consumption expenditures less food and energy price index, which the Fed compares its target to, has averaged roughly 1.5% over the last three years. After strengthening throughout 2016 toward 2%, inflation weakened in the second quarter to 1.5%.
We also looked at other measures of inflation. While the Bureau of Labor Statistics’ alternative inflation measure of CPI less food and energy had surpassed 2% throughout 2016 and into 2017, it shows a similar concerning downward trend. Inflation has declined about 60 basis points to 1.7% in June.
Inflation expectations, which have fallen, are of particular importance, as consumer assumptions about future prices influence purchasing behaviour, and thus economic activity. When consumers expect inflation, they tend to accelerate purchases to get in front of price increases. Likewise, when they expect deflation, they defer purchases, waiting for better prices. As a result, inflation expectations have a significant and somewhat circular impact on economic activity.
Neither market-based nor survey-based expectations paint an optimistic picture of future inflation. Based on implied rates from US Treasuries, the five-year expected inflation rate has fallen below 2%. Rates remain above where they were a year ago, but a continued decline would likely delay future rate hikes. Amid the weakness in 2016, the central bank delivered only one increase compared with its original forecast of four.
Though survey-based inflation expectations have remained relatively steady through 2017, they fail to provide much optimism, either. The University of Michigan's latest consumer survey reported inflation expectations of 2.6%.
In light of this, the FOMC expects near-term inflation to remain somewhat below 2% in the near term, but reiterated expectations that it will stabilise. “Inflation on a 12-month basis is expected to remain somewhat below 2% in the near term but to stabilise around the Committee's 2% objective over the medium term,” the committee said in its June 14 statement.
It appears to us that the market believes weak inflation could slow the pace of Fed rate increases. In turn, this would be positive for gold, as it would limit rising opportunity costs of holding gold.
However, this logic strikes us as a bit circular. With unemployment low, inflation is what’s holding back faster rate hikes. Yet weak inflation means real interest rates are unlikely to be negative. While stronger inflation could reduce real rates, helping gold demand, the FOMC would most likely respond with further rate hikes as inflation neared its target, hurting gold demand. In either case, we don’t see a return to very low or negative real rates that would buoy gold prices.
Furthermore, the real interest rate is still positive, meaning there is still an opportunity cost to holding gold. Rates are not materially different from 2013 levels, when interest in gold investment soured.
While ETF inflows have supported gold prices over the last year or so, they have the potential to flood the market with supply when sentiment turns. As the investment case for gold weakens, ETF holdings are likely to decline, dragging gold prices lower, which could, in turn, trigger further selling.
