How low can gold go?

Of late, the likelihood for a drastic lowering of gold prices (COMEX:GCZ13) has become a common talking-point, with many analysts – ourselves included – believing that a drop to $1,000/oz should not be ruled out in the next year.

Indeed, although the Syria-led rally in oil prices has resulted in a recent upturn in gold prices, the rise is unlikely to be cemented in any longer-term positions. Certainly, a more accurate outlook for gold prices needs to take into account the impacts of developments in three key areas: Investor demand, central bank behavior and trends in the mining industry.

In the past decade (and even more so in the post-crisis environment) these areas have experienced massive shifts, which have largely contributed to the supply and demand disequilibrium that is now of concern. Furthermore, investors still seem wary of gold’s dramatic collapse in the first half of this year. The question has since become, how low should gold be priced in order to bring supply and demand into equilibrium? Let’s try and determine this answer by examining the broader trends currently taking shape.

Investor demand

Whether for bars, coins or exchange-traded products (ETPs), investor demand for gold has historically held steady at around 8%-9% of total demand.

Since the onset of the financial crisis, however – and encouraged by quantitative easing (QE) – investor demand expanded to a high of almost 1,300 tonnes in 2012 (nearly 30% of total demand), helping to propel gold prices to record highs. Undoubtedly, this was an unsustainable rise. And now that U.S. interest rates are rising and the dollar is strengthening – thanks to the Fed’s likely “tapering” of its QE program – investors are relinquishing their holdings of gold.

Indeed, since mid-February, investors have sold 620 tonnes of gold from ETPs. And, just as prices gained momentum on the upside, they are now experiencing momentum on the downside. Selling by investors is helping to push prices lower, further encouraging other investors to capitulate as well. Meanwhile, ETPs, which accounted for 6% of total demand for gold in 2012, have so far this year added net supply equivalent to 14% of last year’s total supply of gold.

When supply and demand of gold is imbalanced, the market can usually rely on natural stabilizers to help the situation, as mentioned earlier. Jewelry demand is one aspect of this mechanism. As prices rise, customers are priced out of the jewelry market – just as when prices fall, consumers can afford to buy more gold.

Global jewelry demand declined significantly between 2008 and 2012, deterred by higher prices. Yet there has been a noticeable increase in jewelry demand in the recent context – seen by the rise in physical premiums for kilo-bars in a range of locations and the backwardation in London prices, which indicates a near-term scarcity of physical metal in the market.

The second key stabilizer is the supply of gold scrap. This is because lower gold prices lead to a sharp drop in global supply of gold scrap. Usually, when the price of gold goes down, we can rely on both of these automatic stabilizers to balance out the equation. But, because of today’s strong outflow from investor demand, it’s tempting to acknowledge that this won’t be enough.

From net demand of around 1,300 tonnes in 2012 (according to GFMS), investment demand has turned into a source of net supply in 2013. If current trends continue, this net supply could reach as much as 400 tonnes or more, with sales from ETPs far outweighing new investment in gold bars. And even if ETP investors stop selling, it will be difficult to see net demand turning positive for the year as a whole.

Central bank behaviour

Certainly, when evaluating the price of gold it is important to keep a close eye on the U.S. and QE. With each announcement from the Fed acting as further confirmation that tapering will begin by the end of this year, interest rates have risen. And higher interest rates means dire news for gold, as it strengthens the dollar and also increases the opportunity-cost of holding gold.

In the decades prior to the financial crisis, central banks acted as a source of supply for gold. Between 1990 and 2009, around 5,600 tonnes of gold were released, mainly by central banks from developed countries. From 2010 onwards, however, the sharply rising price of gold persuaded these central banks to halt further sales of gold.

Since the beginning of the financial crisis, central banks from developing countries have sought to diversify their foreign exchange reserves – making the shift away from the dollar by acquiring increasingly large quantities of gold. Some of these central banks have reached their target holdings over the past 12-18 months, and as a result, are causing a dampening effect on gold prices. This is because they are likely to sell gold at higher prices or buy gold at lower prices – as a means of maintaining a steady proportion of gold within their reserves. International Monetary Fund (IMF) figures show that in the first four months of this year, central banks acquired 160 tonnes of gold. For the year as a whole, we expect net purchases to be at or above 500 tonnes.

Mining industry developments

At current prices, jewelry demand is rising strongly and scrap supply is diminishing. But the enormous decrease in net investment demand needs to be redressed. This is especially the case because solid demand from central banks – which usually plays a key role in driving momentum – is proving insufficient to bring these trends into equilibrium.

What the market needs, therefore, is a fall in mined gold output – meaning that prices should drop to the point at which mining companies close their more marginally profitable operations.

In the gold mining industry, much importance is placed on “all-in” sustaining costs (which among many factors includes costs applicable to sales, exploration expense, advanced projects and R&D). These generally average somewhere between $1,100-1,200/oz – hence our argument that declines in gold prices below those levels will shut in a large portion of global supply. But in the short-run, these are not the right costs to be focusing on.

Instead of long-run average costs – which determine whether mining companies will undertake investment in new ventures – it’s important to consider the marginal costs for each mine, such as cash costs and the net total of by-products. Indeed, it’s these lower marginal costs that will determine whether individual shafts or mines are closed.

Our estimates suggest that – at current prices – the rise in jewelry demand and fall in scrap supply will be able to absorb around 750 tonnes of the decrease in investment in gold. But this leaves an imbalance of as much as 830 tonnes.

To reduce this through lower mine output alone would require prices to fall as low as $800/oz. But of course, if prices were to approach this target we would see an additional rise in jewelry demand and further reductions in scrap supply. Our best guess is that – if selling by investors persists – short-term equilibrium could be reached at the point where gold prices sit somewhere between $1,000/oz and $900/oz.

While we have discounted all-in sustaining costs as an inappropriate measure of the short-term supply curve, over the longer term they are the main determinant of incremental supply. As such, the market will only be able to remain below the industry’s average cost curve so long as demand remains artificially low.

Once investment demand returns to its long-term “normal” of around 8%-9% of the global market, the price level should return to one that compensates mining companies for their continued participation in the gold market.

Not all higher cost mines will reopen, but it would be reasonable to think that prices should stabilize at, or around, the long-run costs of production for the vast majority of the gold industry. This suggests that a reasonable medium-term target for gold prices would be somewhere in the vicinity of $1,150-1,250/oz.

Recognizing this long-run equilibrium, the market may be rational enough to halt its breakneck decline as investors resist selling further volumes of gold below the long-run cost curve. With many investors remaining trapped in a falling market, however, there may be more rounds of ‘blood-letting’ to go before some degree of sanity returns.

About the Author

Bernard Dahdah is responsible for precious metals research at Natixis and has been covering the gold, silver platinum and palladium markets for four years.  After various roles at Merrill Lynch and HSBC, Bernard pursued his studies at the Cass business school in London where