Agoracom Blog

Gold’s Outlook for 2020 – SPONSOR: American Creek Resources $AMK.ca $TUD.ca $SII.ca $GTT.ca $AFF.ca $SEA.ca $SA $PVG.ca $AOT.ca

Posted by AGORACOM at 2:20 PM on Friday, January 10th, 2020

This article is an overview of the economic conditions that will drive the gold price in 2020 and beyond. The turn of the credit cycle, the effect on government deficits and how they are to be financed are addressed.

In the absence of foreign demand for new US Treasuries and of a rise in the savings rate the US budget deficit can only be financed by monetary inflation. This is bound to lead to higher bond yields as the dollar’s falling purchasing power accelerates due to the sheer quantity of new dollars entering circulation. The relationship between rising bond yields and the gold price is also discussed.

It may turn out that the recent extraordinary events on Comex, with the expansion of open interest failing to suppress the gold price, are an early recognition in some quarters of the US Government’s debt trap.

The strains leading to a crisis for fiat currencies are emerging into plain sight.

Introduction

In 2019, priced in dollars gold rose 18.3% and silver by 15.1%. Or rather, and this is the more relevant way of putting it, priced in gold the dollar fell 15.5% and in silver 13%. This is because the story of 2019, as it will be in 2020, was of the re-emergence of fiat currency debasement. Particularly in the last quarter, the Fed began aggressively injecting new money into a surprisingly illiquid banking system through repurchase agreements, whereby banks’ reserves at the Fed are credited with cash loaned in return for T-bills and coupon-bearing Treasuries as collateral. Furthermore, the ECB restarted quantitative easing in November, and the Bank of Japan stands ready to ease policy further “if the momentum towards its 2% inflation target comes under threat” (Kuroda – 26 December).

The Bank of Japan is still buying bonds, but at a pace which is expected to fall beneath redemptions of its existing holdings. Therefore, we enter 2020 with money supply being expanded by two, possibly all three of the major western central banks. Besides liquidity problems, the central bankers’ nightmare is the threat that the global economy will slide into recession, though no one will confess it openly because it would be an admission of policy failure. And policy makers are also terrified that if bankers get wind of a declining economy, they will withdraw loan facilities from businesses and make things much worse.

Of the latter concern central banks have good cause. A combination of the turn of the credit cycle towards its regular crisis phase and Trump’s tariff war has already hit international trade badly, with exporting economies such as Germany already in recession and important trade indicators, such as the Baltic dry index collapsing. No doubt, President Trump’s most recent announcement that a trade deal with China is ready for signing is driven by an understanding in some quarters of the White House that over trade policy, Trump is turning out to be the turkey who voted for Christmas. But we have heard this story several times before: a forthcoming agreement announced only to be scrapped or suspended at the last moment.

The subject which will begin to dominate monetary policy in 2020 is who will fund escalating government deficits. At the moment it is on few investors’ radar, but it is bound to dawn on markets that a growing budget deficit in America will be financed almost entirely by monetary inflation, a funding policy equally adopted in other jurisdictions. Furthermore, Christine Lagarde, the new ECB president, has stated her desire for the ECB’s quantitative easing to be extended from government financing to financing environmental projects as well.

2020 is shaping up to be the year that all pretence of respect for money’s role as a store of value is abandoned in favour of using it as a means of government funding without raising taxes. 2020 will then be the year when currencies begin to be visibly trashed in the hands of their long-suffering users.

Gold in the context of distorted markets

At the core of current market distortions is a combination of interest rate suppression and banking regulation. It is unnecessary to belabour the point about interest rates, because minimal and even negative rates have demonstrably failed to stimulate anything other than asset prices into bubble territory. But there is a woeful lack of appreciation about the general direction of monetary policy and where it is headed.

The stated intention is the opposite of reality, which is not to rescue the economy: while important, from a bureaucrat’s point of view that is not the greatest priority. It is to ensure that governments are never short of funds. Inflationary financing guarantees the government will always be able to spend, and government-licenced banks exist to ensure the government always has access to credit.

Unbeknown to the public, the government licences the banks to conduct their business in a way which for an unlicensed organisation is legally fraudulent. The banks create credit or through their participation in QE they facilitate the creation of base money out of thin air which is added to their reserves. It transfers wealth from unsuspecting members of the public to the government, crony capitalists, financial speculators and consumers living beyond their means. The government conspires with its macroeconomists to supress the evidence of rising prices by manipulating the inflation statistics. So successful has this scheme of deception been, that by fuelling GDP, monetary debasement is presented as economic growth, with very few in financial mainstream understanding the deceit.

The government monopoly of issuing money, and through their regulators controlling the expansion of credit, was bound to lead to progressively greater abuse of monetary trust. And now, in this last credit cycle, the consumer who is also the producer has had his income and savings so depleted by continuing monetary debasement that he can no longer generate the taxes to balance his government’s books later in the credit cycle.

The problem is not new. America has not had a budget surplus since 2001. The last credit cycle in the run up to the Lehman crisis did not deliver a budget surplus, nor has the current cycle. Instead, following the Lehman crisis we saw a marked acceleration of monetary inflation, and Figure 2 shows how dollar fiat money has expanded above its long-term trend since then.

In recent years, the Fed’s attempt to return to monetary normality by reducing its balance sheet has failed miserably. After a brief pause, the fiat money quantity has begun to grow at a pace not seen since the immediate aftermath of the Lehman crisis itself and is back in record territory. Figure 1 is updated to 1 November, since when FMQ will have increased even more.

In order to communicate effectively the background for the relationship between gold and fiat currencies in 2020 it is necessary to put the situation as plainly as possible. We enter the new decade with the highest levels of monetary ignorance imaginable. It is a systemic issue of not realising the emperor has no clothes. Consequently, markets have probably become more distorted than we have ever seen in the recorded history of money and credit, as widespread negative interest rates and negative-yielding bonds attest. In our attempt to divine the future, it leaves us with two problems: assessing when the tension between wishful thinking in financial markets and market reality will crash the system, and the degree of chaos that will ensue.

The timing is impossible to predict with certainty because we cannot know the future. But, if the characteristics of past credit cycles are a guide, it will be marked with a financial and systemic crisis in one or more large banks. Liquidity strains suggest that event is close, even within months and possibly weeks. If so, banks will be bailed, of that we can be certain. It will require central banks to create yet more money, additional to that required to finance escalating government budget deficits. Monetary chaos promises to be greater than anything seen heretofore, and it will engulf all western welfare-dependent economies and those that trade with them.

We have established that between keeping governments financed, bailing out banks and perhaps investing in renewable green energy, the issuance of new money in 2020 will in all probability be unprecedented, greater than anything seen so far. It will lead to a feature of the crisis, which may have already started, and that is an increase in borrowing costs forced by markets onto central banks and their governments. The yield on 10-year US Treasuries is already on the rise, as shown in Figure 3.


Assuming no significant increase in the rate of savings and despite all attempts to suppress the evidence, the acceleration in the rate of monetary inflation will eventually lead to runaway increases in the general level of prices measured in dollars. As Milton Friedman put it, inflation [of prices] is always and everywhere a monetary phenomenon.

Through QE, central banks believe they can contain the cost of government funding by setting rates. What they do not seem to realise is that while to a borrower interest is a cost to set against income, to a lender it reflects time-preference, which is the difference between current possession, in this case of cash dollars, and possession at a future date. Unless and until the Fed realises and addresses the time preference problem, the dollar will lose purchasing power. Not only will it be sold in the foreign exchanges, but depositors will move to minimise their balances and creditors their ownership of debt.

If, as it appears in Figure 3, dollar bond yields are beginning a rising trend, the inexorable pull of time preference is already beginning to apply and further rises in bond yields will imperil government financing. The Congressional Budget Office assumes the average interest rate on debt held by the public will be 2.5% for the next three years, and that net interest in fiscal 2020 will be $390bn, being about 38% of the projected deficit of $1,008bn. Combining the additional consequences for government finances of a recession with higher bond yields than the CBO expects will be disastrous.

Clearly, in these circumstances the Fed will do everything in its power to stop markets setting the cost of government borrowing. But we have been here before. The similarities between the situation for the dollar today and the deterioration of British government finances in the early to mid-1970s are remarkable. They resulted in multiple funding crises and an eventual bail-out from the IMF. Except today there can be no IMF bail-out for the US and the dollar, because the bailor gets its currency from the bailee.

Nearly fifty years ago, in the UK gold rose from under £15 per ounce in 1970 to £80 in December 1974. The peak of the credit cycle was at the end of 1971, when the 10-year gilt yield to maturity was 7%. By December 1974, the stock market had crashed, a banking crisis had followed, price inflation was well into double figures and the 10-year gilt yield to maturity had risen to over 16%.

History rhymes, as they say. But for historians the parallels between the outlook for the dollar and US Treasury funding costs at the beginning of 2020, and what transpired for the British economy following the Barbour boom of 1970-71 are too close to ignore. It is the same background for the relationship between gold and fiat currencies for 2020 and the few years that follow.

Gold and rising interest rates

Received investment wisdom is that rising interest rates are bad for the gold price, because gold has no yield. Yet experience repeatedly contradicts it. Anyone who remembers investing in UK gilts at a 7% yield in December 1971 only to see prices collapse to a yield of over 16%, while gold rose from under £15 to £80 to the ounce over the three years following should attest otherwise.

Part of the error is to believe that gold has no yield. This is only true of gold held as cash and for non-monetary usage. As money, it is loaned and borrowed, just like any other form of money. Monetary gold has its own time preference, as do government currencies. In the absence of state intervention, time preferences for gold and government currencies are set by their respective users, bearing in mind the characteristics special to each. It is not a subject for simple arbitrage, selling gold and buying government money to gain the interest differential, because the spread reflects important differences which cannot be ignored. It is like shorting Swiss francs and buying dollars in the belief there is no currency risk.

The principal variable between the time preferences of gold and a government currency is the difference between an established form of money derived from the collective preferences of its users, for which there is no issuer risk, and state-issued currency which becomes an instrument of funding by means of its debasement.

The time preference of gold will obviously vary depending on lending risk, which is in addition to an originary rate, but it is considerably more stable than the time preference of a fiat currency. Gold’s interest rate stability is illustrated in Figure 4, which covers the period of the gold standard from the Bank Charter Act of 1844 to before the First World War, during which time the gold standard was properly implemented. With the exception of uncontrolled bank credit, sterling operated as a gold substitute.


Admittedly, due to problems created by the cycle of bank credit, these year-end values conceal some significant fluctuations, such as at the time of the Overend Gurney collapse in 1866 when borrowing rates spiked to 10%. The depression following the Barings crisis of 1890 stalled credit demand which is evident from the chart. However, wholesale borrowing rates, which were effectively the cost of borrowing in gold, were otherwise remarkably stable, varying between 2-3½%. Some of this variation can be ascribed to changing perceptions of general borrower risk and some to changes in industrial investment demand, related to the cycle of bank credit.

Compare this with dollar interest rates since 1971, when the dollar had suspended the remaining fig-leaf of gold backing, which is shown in Figure 5 for the decade following.


In February 1972 the Fed Funds rate was 3.29%, rising eventually to over 19% in January 1981. At the same time gold rose from $46 to a high of $843 at the morning fix on 21 January 1980. Taking gold’s originary interest rate as approximately 2% it required a 17% interest rate penalty to dissuade people from hoarding gold and to hold onto dollars instead.

In 1971, US Government debt stood at 35% of GDP and in 1981 it stood at 31%. The US Government ran a budget surplus over the decade sufficient to absorb the rising interest cost on its T-bill obligations and any new Treasury funding. America enters 2020 with a debt to GDP ratio of over 100%. Higher interest rates are therefore not a policy option and the US Government, and the dollar, are ensnared in a debt trap from which the dollar is unlikely to recover.

The seeds of the dollar’s destruction were sown over fifty years ago, when the London gold pool was formed, whereby central banks committed to help the US maintain the price at $35, being forced to do so because the US could no longer supress the gold price on its own. And with good reason: Figure 6 shows how the last fifty years have eroded the purchasing power of the four major currencies since the gold pool failed.

 

Over the last fifty years, the yen has lost over 92%, the dollar 97.6%, the euro (and its earlier components 98.2% and sterling the most at 98.7%. And now we are about to embark on the greatest increase of global monetary inflation ever seen.

The market for physical gold

In recent years, demand for physical gold has been strong. Chinese and Indian private sector buyers have to date respectively accumulated an estimated 17,000 tonnes (based on deliveries from Shanghai Gold Exchange vaults) and about 24,000 tonnes (according to WGC Director Somasundaram PR quoted in India’s Financial Express last May).

It is generally thought that higher prices for gold will deter future demand from these sources, with the vast bulk of it being categorised as simply jewellery. But this is a western view based on a belief in objective values for government currencies and subjective prices for gold. It ignores the fact that for Asians, it is gold that has the objective value. In Asia gold jewellery is acquired as a store of value to avoid the depreciation of government currency, hoarded as a central component of a family’s long-term wealth accumulation.

Therefore, there is no certainty higher prices will compromise Asian demand. Indeed, demand has not been undermined in India with the price rising from R300 to the ounce to over R100,000 today since the London gold pool failed, and that’s despite all the government disincentives and even bans from buying gold.

Additionally, since 2008 central banks have accumulated over 4,400 tonnes to increase their official reserves to 34,500 tonnes. The central banks most active in the gold market are Asian, and increasingly the East and Central Europeans.

There are two threads to this development. First there is a geopolitical element, with Russia replacing reserve dollars for gold, and China having deliberately moved to control global physical delivery markets. And second, there is evidence of concern amongst the Europeans that the dollar’s role as the reserve currency is either being compromised or no longer fit for a changed world. Furthermore, the rising power of Asia’s two hegemons continues to drive over two-thirds of the world’s population away from the dollar towards gold.

Goldmoney estimates there are roughly 180,000 tonnes of gold above ground, much of which cannot be categorised as monetary: monetary not as defined for the purposes of customs reporting, but in the wider sense to include all bars, coins and pure gold jewellery accumulated for its long-term wealth benefits through good and bad times. Annual mine production adds 3,000-3,500 tonnes, giving a stock to flow ratio of over 50 times. Put another way, the annual increase in the gold quantity is similar to the growth in the world’s population, imparting great stability as a medium of exchange.

These qualities stand in contrast to the increasingly certain acceleration of fiat currency debasement over the next few years. Anyone prepared to stand back from the financial coalface can easily see where the relationship between gold and fiat currencies is going. Most of the world’s population is moving away from the established fiat regime towards gold as a store of value, their own fiat currencies lacking sufficient credibility to act as a dollar alternative. And financial markets immersed in the fiat regime have very little physical gold in possession. Instead, where it is now perceived that there is a risk of missing out on a rise in the gold price, investors have begun accumulating in greater quantities the paper alternatives to physical gold: ETFs, futures, options, forward contracts and mining shares.

Paper markets

From the US Government’s point of view, gold as a rival to the dollar must be quashed, and the primary purpose of futures options and forwards is to expand artificial supply to keep the price from rising. In a wider context, the ability to print synthetic commodities out of thin air is a means of suppressing prices generally and we must not be distracted by claims that derivatives improve liquidity: they only improve liquidity at lower prices.

When the dollar price of gold found a major turning point on 17 December 2015, open interest on Comex stood at 393,000 contacts. The year-end figure today is nearly double that at 786,422 contracts, representing an increase of paper supply equivalent to 1,224 tonnes. But that is not all. Not only are there other regulated derivative exchanges with gold contracts, but also there are unregulated over the counter markets. According to the Bank for International Settlements from end-2015 unregulated OTC contracts (principally London forward contracts) expanded by the equivalent of 2,450 tonnes by last June, taken at contemporary prices. And we must not forget the unknown quantity of bank liabilities to customers’ unallocated accounts which probably involve an additional few thousand tonnes.

In recent months, the paper suppression regime has stepped up a gear, evidenced by Comex’s open interest rising. This is illustrated in Figure 7.


There are two notable features in the chart. First, the rising gold price has seen increasing paper supply, which we would expect from a market designed to keep a lid on prices. Secondly instead of declining with the gold price, open interest continued to rise following the price peak in early September while the gold price declined by about $100. This tells us that the price suppression scheme has run into trouble, with large buyers taking the opportunity to increase their positions at lower prices.

In the past, bullion banks have been able to put a lid on prices by creating Comex contracts out of thin air. The recent expansion of open interest has failed to achieve this objective, and it is worth noting that the quantity of gold in Comex vaults eligible for delivery and pledged is only 2% of the 2,446-tonne short position. In London, there are only 3,052 tonnes in LBMA vaults (excluding the Bank of England), which includes an unknown quantity of ETF and custodial gold. Physical liquidity for the forward market in London is therefore likely to be very small relative to forward deliveries. And of course, the bullion banks in London and elsewhare do not have the metal to cover their obligations to unallocated account holders, which is an additional consideration.

Clearly, there is not the gold available in the system to legitimise derivative paper. It now appears that paper gold markets could be drifting into systemic difficulties with bullion banks squeezed by a rising gold price, short positions and unallocated accounts.

There are mechanisms to counter these systemic risks, such as the ability to declare force majeure on Comex, and standard unallocated account contracts which permit a bullion bank to deliver cash equivalents to bullion obligations. But the triggering of any such escape from physical gold obligations could exacerbate a buying panic, driving prices even higher. It leads to the conclusion that any rescue of the bullion market system is destined to fail.

A two-step future for the gold price

It has been evident for some time that the world of fiat currencies has been drifting into ever greater difficulties of far greater magnitude than can be contained by spinning a few thousand tonnes of gold back and forth on Comex and in London. That appears to be the lesson to be drawn from the inability of a massive increase in open interest on Comex to contain a rising gold price.

It will take a substantial upward shift in the gold price to appraise western financial markets of this reality. In combination with systemic strains increasing, a gold price of over $2,000 may do the trick. Professional investors will have found themselves wrongfooted; underinvested in ETFs, gold mines and regulated derivatives, in which case their gold demand is likely to drive one or more bullion houses into considerable difficulties. We might call this the first step in a two-step monetary future.

The extent to which gold prices rise could be substantial, but assuming the immediate crisis itself passes, banks having been bailed in or out, and QE accelerated in an attempt to put a lid on government bond yields, then the gold price might be deemed to have risen too far, and due for a correction. But then there will be the prospect of an accelerating loss of purchasing power for fiat currencies as a result of the monetary inflation, and that will drive the second step as investors realise that what they are seeing is not a rising gold price but a fiat currency collapse.

The high levels of government debt today in the three major jurisdictions appear to almost guarantee this outcome. The amounts involved are so large that today’s paper gold suppression scheme is likely to be too small in comparison and cannot stop it happening. The effect on currency purchasing powers will then be beyond question. Monetary authorities will be clueless in their response, because they have all bought into a form of economics that puts what will happen beyond their understanding.

As noted above, the path to a final crisis for fiat currencies might have already started, with the failure by the establishment to suppress the gold price through the creation of an extra 100,000 Comex contracts. If not, then any success by the monetary authorities to reassert control is likely to be temporary.

Perhaps we are already beginning to see the fiat currency system beginning to unravel, in which case those that insist gold is not money will find themselves impoverished.

Source: Goldmoney Insights

https://www.goldmoney.com/research/goldmoney-insights/gold-s-outlook-for-2020

Tags: , , , , , , , , , , , , , , ,

Comments are closed.