Economic explanations for gold’s sharp fall

In the weekly Buttonwood blog published in the previous week’s edition of The Economist on April 20, the topic was the plunge of gold earlier this month. We agree that it is hard to find convincing economic explanations for the sharp fall. However, Buttonwood is wrong in some of the conclusions about economic developments and the price of gold.

The Economist states rightly that not only gold, but commodity prices in general had been falling year-to-date. Crude oil prices also were down by mid-April, but WTI is trading again at the level of the start of this year, while Brent is still down 8.3% YTD despite the recent recovery. Also some base metals reached multi-year lows earlier this month. Especially copper, which is often dubbed the metal with a PhD. in economics and regarded as a bellwether for the global economic development, is not reflecting forecasts of a rebounding world economy. Also grain prices have come down compared with the level prevailing at the end of the harvest season in the Northern hemisphere last year. The development of commodity prices is certainly not underlining the optimism, which had led to a shift from gold into equities being observed during the first quarter. However, the decline of prices for Brent oil and the grains has an impact on headline CPI inflation rates because they are usually highly weighted in consumer price indices. Thus, the decline of commodity prices has reduced the risk of rising headline inflation despite the aggressive monetary easing by major central banks, which increased their balance sheets substantially over the last couple of years.

Buttonwood argues further “… if economic sentiment were improving significantly you would expect investors to sell government bonds as well as gold. Here Buttonwood errs. If the global economy would expand strongly, then the demand for commodities would also increase. One reason for the weaker oil prices were the downward revisions for oil demand by OPEC, EIA and IEA due to slower than previously expected global growth. Thus, an increase of headline inflation would have to be expected. In addition, Japan targets now a CPI inflation rate of 2% over the next two years to overcome the depression. Also the Fed forecasts that the extremely lower interest rates will prevail until around mid-2015. Thus, in major economies, monetary policy remains highly accommodative. Furthermore, the risk would increase that the balance sheet extensions would lead to increasing lending to the private sector and that also core inflation would edge up if global growth accelerates. Therefore, one would expect that gold would be bought and bonds would be sold if economic sentiment were improving significantly. Thus, the phenomenon observed since mid-March that bond yields declined and gold dropped just reflects that short-term inflation risks have declined, which is also mirrored in the development of inflation expectations measured by the spread between nominal and inflation-linked US Treasury paper.

The Fed clearly stated conditions under which QE would be stopped. These conditions are not negative for gold because either inflation has to rise above a threshold or the unemployment rate has to drop to a target level. The first case is obviously positive for gold as it would be bought as a hedge against increasing inflation rates. But also the case of terminating QE on unemployment rate falling to 6% would be positive for gold as it requires stronger economic growth and would lead to higher wages, which one would expect to translate into higher core inflation.

Buttonwood argues further that gold were hard to value because of not having a yield or earnings. Certainly, gold has no earnings. But this is also the case with many companies, especially start-ups, which even produce losses during the first few years. Nevertheless, the stock market prices the equities of those companies, once they are publicly listed. Also for bonds with a fixed nominal coupon, the market has to find a fair yield to reflect the fundamentals. Negative real yields on 10yr notes, as it is the case for US Treasuries or German Bunds are certainly not a fair value for investors.

For gold, as for any other good, marginal production costs are one yardstick for longer-term pricing. However, quantitative methods also allow the development of various fair value models. Whether those models are based on traditional ordinary least square (OLS) methods or vector autoregressive (VAR) methods, they provide a fair value for the price of gold. Also volatility models, which include fundamental factors for explaining the return on gold and its volatility (GARCH-X models), could be applied. They all show that there are three major factors play a role for explaining the price moves of gold. These factors are the stock market, the price of crude oil and the US dollar index. While the US dollar index and the oil price were negative for gold, the stock market – represented by the S&P 500 index – was positive. However, the two negative factors were not sufficient to explain the plunge of gold and other precious metals in mid-April.

That was not super, Mario!

According to media reports, a statement by ECB president Mario Draghi that Cyprus would sell a major part of its gold reserves to raise about 400 mil euros (as a part of the bailout package agreed between the troika and Cyprus’ president) was the trigger for the massive sell-off in precious metals. This was not super for several reasons.

Silence is golden. This is not only the title of a song performed by a group called the Marmalades when Mr. Draghi was a teenager, but also an old adage. The volume of total gold sales intended by Cyprus is only the turnover in the gold market of a few days. Nevertheless, in every market, the announcement of a larger sale leads immediately to weaker prices. Thus, Mr. Draghi’s statement reduced the potential revenues Cyprus might obtain by more than 230$/oz within two trading days. Therefore, the few words of the ECB presidents reduced the expected proceeds for Cyprus by almost 15%. This was really not a super help for Cyprus. Mario Draghi better had kept his mouth shut.

The gold market is a place, where some market participants believe in conspiracy theories. To make it clear, we are not convinced by those theories. Nevertheless, the plunge of the gold price triggered by the statement of Mr. Draghi revived the conspiracy theory that this comment was part of an orchestrated action by central banks to manipulate the market. Central banks were also supposed to be the force behind selling 400 tons of gold on Friday April 12 at Comex. On analyst even accused central banks to be the dark force responsible for declining prices of gold on some days either at the opening of the pit session at Comex or the afternoon fixing in London. The argument by the gold bugs is that a rising gold price would express distrust in paper money and therefore, central banks would manipulate the price down to signal that fiat money could be trusted and that gold would not be a safe haven. However, if major central banks would really sell gold to push prices lower, it would be reflected in their balance sheets. And there is not any hint for market manipulation.

Gold held by the central banks are part of their reserves. Any disposition about the management of these reserves is in the discretion of the governing body of the central bank. In the eurozone, the ECB and the national central banks are independent and should not set under any political pressure, neither by the EU nor national governments. This is guaranteed by the EU treaties. However, in the case of Cyprus, the insistence of the EU on selling gold reserves to obtain the requested contribution of Cyprus to the bailout is a clear violation of the independence of central banks in the eurozone. What is worrisome is that the ECB did not protest against this demand but Mr. Draghi even gave his approval.

That the plan of Cyprus to obtain 400 mil euros from the sale of a part of its gold reserves led to such a steep plunge of gold and other precious metal prices is also the result of the mismanagement of the debt crisis by European politicians. Too often, politicians, in particular the German finance minister, promised first that a certain measure would not be taken, later then it was only in a special case before it became normal procedure. Thus, all trust had been destroyed. Therefore, it really does not come as a surprise that the market immediately speculated that also Ireland and the southern European countries might be forced by the EU to sell gold reserves. And the gold holdings of Spain and Italy are far higher than those of Cyprus.

If a market plunges, the usual suspects are the infamous hedge funds shorting to market to make a quick profit. This was also a suspicion some commentators voiced after the plunge continued on Monday this week. At a first glance, the record volume of gold futures traded at Comex on Monday at more than 750K contracts in the front month June contract appeared to confirm this suspect. However, the latest CFTC report on the “Commitment of Traders” surprised. Large speculators even increased their net long position in Comex gold futures in the week ending April 16 to 128,882 contracts, a plus of 9,523 contracts. Furthermore, if hedge funds really would massively short gold futures, then one would expect a rise of open interest. However, the total open interest declined by almost 2,500 to 413,083 contracts.

The increase of the net long position held by large speculators, the physical buying from retail investors and commercial demand in Asia after the plunge, these all are positive indications. Furthermore, the market was oversold and technical indicators returned from the oversold back into the neutral zone, which is normally a buy signal. Thus, there are good chances for a recovery of gold and other precious metals. However, the sentiment, especially among analysts at major investment banks, is still negative and outflows from gold ETFs continued. Therefore, the risks to the downside remain considerable.     

Black Friday for precious metals

Until the middle of last week, it still looked like gold might manage to stay within the sideways trading range. However, on Friday, gold plunged by 87$/oz or by 5.6%. Thus, gold has retraced the entire advance from the low at July 1, 2011 at 1,478$/oz to the all time high at 1,920$/oz. With gold having fallen by around 23% from the high, the market is now regarded by many technical analysts as being in a bear market. The plunge of gold also dragged other precious metals lower.

In media reports, two factors were stated as being responsible for the black Friday in precious metals: economic woes and Cyprus.  Ahead of the meeting of European finance ministers on Friday, Cyprus stated that the financing needs under the EU/IMF bailout have risen to 23bn euro due to the deterioration of the economy, which will depress revenues. Cyprus is another case demonstrating that the remedy prescribed by the troika is not heeling the patient. Austerity is not improving the economic situation but only makes a recession worse. Austerity leads to even bigger holes in the budget.

Normally, this would not have caused such a strong negative price reaction in the gold market. However, there were also reports, that Cyprus would have to sell 0.4 bn euro worth of gold, which are most the country’s gold reserves. While this had not been confirmed officially, it had nevertheless an impact on gold. But the intraday chart shows that gold reacted at the start of trading in Europe and then moved sideways above the 1,540$/oz level. Also the London AM fixing at 1,548$/oz was only 7$ lower than the price of the morning fixing at the day before.

The second factor quoted was economic woes. Certainly, some economic figures came in lower than expected.  However, the University of Michigan consumer sentiment index is subject to considerable revisions, as the recent report demonstrated. Working day effects of the Easter holidays are hard to eliminate by the usual seasonal adjustment procedures. Thus, data for the months of March and April should always be taken with some caution. For example, also last year, the number of new jobs in March came in much lower than expected and triggered a rally in the US Treasury market. However, this figure had been revised up and also the annual benchmark revisions led to another upward revision.  

Thus, economic woes might be overdone and not justified. However, the crucial question is, do they really explain the plunge of precious metals prices. In various econometric models for the gold price, there are three fundamental factors, which play a significant role – the US stock market, crude oil and the US dollar index. Economic woes should be reflected in falling stock markets. However, the S&P 500 index posted another rise last week. Thus, it appears that economic woes don’t move the stock market. But it would be foolish to sell gold on concerns about the economic outlook and to invest the proceeds in the stock market. Crude oil and the US dollar index were only modestly lower in the week over week comparison and thus, both together do not support such a strong plunge of the gold price.

There were reports that the Merrill Lynch division of Bank of America sold 4 million ounces of gold at the opening on Comex. This fits more with the intraday price development of gold on Friday. The value of such a transaction is about 6bn USD, far more than the amount Cyprus is expected to raise by selling the major part of its gold reserves. Did I miss Bart Chilton stating that the CFTC would start investigating gold price manipulations by US banks at the Comex division of the CME?

After the drop of precious metals prices last Friday, the market is vulnerable for further losses as the sentiment turned bearish and investors might sell further precious metals. But on the other hand, the market is already oversold. Thus, also stabilization can not be ruled out. However, for the time being, the risk appears to be more biased to the downside.

Is gold really in a new multi-year bear market? No!

Reuters reported last week that technical analysts regard gold as being in a new multi-year bear market. Reporting from a conference of the US Market Technicians Association, a Reuters’ journalist interviewed Robert Prechter, the head of Elliott Wave International. Mr. Prechter is quoted as saying that gold would be in a major bear market lasting for some years. Furthermore, Mr. Prechter compared the technical situation of gold with the development of the NASDAQ Composite index at the end of the dot-com bubble and the real-estate boom in the US in 2006. Other technical analysts also expressed bearish views for the price of gold and expect that gold had seen a multi-year high.

It is correct that it is now almost 18 months ago that gold had reached its all-time high at 1,920$/oz in early September 2011. It is also correct that gold is trading now for some time below the steep upward trend line, which originates at the low made in 2008. However, is the break of an upward trend line already a proof that a market is now in a bearish trend? In the binary world of the computer age, one might be inclined to answer this question with a yes. But this is not correct for several reasons.

First, in the literature on technical analysis, there is a broad consensus, that there are three types of a trend, an upward, a downward and a sideways trend. Thus, breaking a trend line does not automatically lead to a bearish trend. Second, according to the Dow Theory, a trend is only reversed if prices fall below the pivot low of the move leading to the high. This is not the case, as gold still trades above 1,478 $/oz, the low made on July 1, 2011.

Third, the selection of the starting point for a trend is often highly subjective. Gold hit a major low in the summer of 1998 at 251$/oz in the wake of the Russian debt crisis. Another low in 2001 came very close to this low. If one dates the low in February 2001 as the starting point for the trend, then the upward trend is still intact. Also using correction lows between this date and the summer of 2006 as starting points would lead to trend lines coming on close to the long-term trend line from 2001. This leads to the fourth point. It could be observed very often that steep trend lines are broken, but trend lines with a lower slope remain well in place and that prices do not retrace back to those lines.

Furthermore, it is also not appropriate to compare the current situation of the gold market with bubbles in stock markets and the US real-estate market in the mid-2000^th. Whether it was the Japanese Nikkei index in 1990 or the NASDAQ index ten years later the bursting of a bubble showed the same pattern in the charts. Stock prices fell sharp and quickly. The NASDAQ index lost more than 40% from its high in 2000 within less than 3 months. After a short rebound, the NASDAQ index lost almost 75% from the high within 18 month. The development of US housing prices after the peek in 2006 showed a similar pattern. Also stock markets in 2008 showed a similar pattern of sharp losses within only a relatively short period of time. However, looking at the chart of gold, this pattern did not emerge.

Another technical tool to compare the current situation of gold with stock markets after bursting bubbles is to look at the Fibonacci retracements. Unlike expressing losses as a percentage of the highest price reached, this tool is looking at the percentage relationship between a retracement and the preceding movement. An upward move is regarded to remain intact, if the retracement does not reach certain Fibonacci levels. The NASDAQ in 2000 retraced more than the 50% Fibonacci level within 3 months and 100% of the preceding upward move within 18 months. In the case of gold, the retracement since the high in 2011 has been made remained always above the first target level of the 38.2% Fibonacci level, which is at 1,446$/oz.

Despite the sharp decline last week, gold is still in a sideways trading range. And as long as gold trades above the low from July 1, 2011 at 1,478$/oz, we regard gold as remaining stuck in a sideways trend. The sentiment for gold is bearish among analysts and also some investors, as the development of ETF holdings and the CoT report of the CFTC show. Thus, the risk of a further downward move is rather high. However, technical indicators point to an oversold market in the daily and weekly time frame. Therefore, gold has still a good chance to remain further in the almost 18 months old trading range. Thus, gold is not in a new multi-year downward trend, at least not yet.