Gold, Global M2 and Central Bank Reserves

The recent plunge in the price of gold has understandably caused quite a stir. We offer some observations about it in light of the role gold plays in central bank reserves. We then put the recent moves in historical perspective with a bit of technical analysis, and finally we present gold as financial catastrophe/inflation insurance.

Gold and Global M2

We know that, globally, M2 has increased over time. Central bank reserves have kept pace, as shown below.

Has the gold component of those reserves kept pace? Below is a chart that depicts the ratio of gold reserves to total reserves. It has held relatively steady over the period shown, suggesting a certain discipline on the part of central banks to maintain a steady hand in the reserves department (the pink curve).

Gold and Central Bank Reserves

But look again. Gold has risen significantly since 1999 - from $276 to $1598 per ounce. If we hold the value of gold steady from the beginning of the period we see that the absolute ratio of gold to total reserves declines very rapidly over the ten years shown (the blue curve). In other words, as M2 has grown, all reserves including gold have grown in dollar terms. However, the physical ratio of gold to all reserves has certainly not grown in proportion; in fact it declined by 6.54% while its market value increased by 416%.

The practice of measuring the gold reserves ratio in current dollars has some interesting effects, one being that the ratio is a function of the price of gold. As we have just seen, if increases in M2 (and presumably reserves) by market action drive up the price of gold the ratio increases by definition - thereby masking inflation, perhaps enabling it. However, this effect by itself apparently isn't enough to maintain the ratio, for "the first quarter of 2013 marked the seventh consecutive quarter of net gold purchases by central banks," according to the latest Gold Demands Report by the World Gold Council.

The graph below illustrates: the market value of gold reserves (and thus, the gold reserves ratio) increases even as physical reserves decrease (the blue curve). Then, in the period beginning in 2008, as the nominal ratio bounces off 9% (see second graph), physical gold reserves begin increasing even as the market price of gold does.

Another effect of a dollar-weighted gold reserves ratio is more appropriate to current conditions: as gold prices fall the ratio falls proportionately. At the end of the last data period depicted, assuming no subsequent M2 or total reserves growth, the gold reserves ratio would have declined from 12% to 9% based strictly upon a decline in the price of gold from $1598 to $1200. Thus falling gold prices would force central banks to acquire gold in order to maintain the ratio, should that be their mandate. A look at the pink curve on the second graph above shows that, over the period considered, the gold reserves ratio has in fact been roughly in a 9-12% range.

In absolute terms the gold reserves ratio has been decreasing steadily and amounts to continuous debasement of the currency. As depicted, even if M2 growth were halted the ratio in 1999 dollar terms would still be sitting at a low, suggesting the possibility of realignment. Under current policy and economic conditions - that of aggressive stimulus and little economic growth -- either because of central bank purchases or by currency inflation, the price of gold must increase on average, but for this caveat:

Gold is, however, particularly subject to short-term price fluctuations. Nation-states and their central banks hold about half of the globe's aboveground stores of gold, with which they can and do engineer extreme short-term price declines. And, as more than 95 percent of the gold mined in the past still exists, with most of it in bullion bar form, the globe's stores of gold are perhaps a hundred times larger than annual world production of gold - and nearly a hundred times more potentially influential, on the supply side, to short term price movements.[1]

This would make the business of timing one's gold purchases a very tricky one indeed. While the numbers and ratios in the above citation are as of 1979, there is no reason to conclude that the principal point is less valid today. In fact, as we noted above and elsewhere, central banks were net buyers of gold throughout the Q1 2013, while the commodities and related markets recorded historic price declines. 

Notes on the Recent Price Action in Gold

Adding one last dimension to this discussion, the recent price action of gold is put in perspective by showing common and actual price retracement amounts in the table below for both the spot market and GLD shares.

The top two rows show peak and current prices. Below that are estimated "breakout" points for the given year, the breakout-to-peak range, and the prices corresponding to the common 50% and 67% retracement amounts. In the final row is the actual retracement as of this writing. (sources: World Gold Council, SPDR Gold Shares; retracements: calculated).

Along with a historical price perspective, we have seen data suggesting that, in an environment of increasing M2, central banks acquire gold regardless of whether it is rising or falling in price. We have seen evidence in our data sample (1999 - 2011) of the "gold reserves ratio" remaining in a range between roughly 9% and 12%, and estimated that, had current market prices prevailed at the time, all other variables being equal, the ratio would be at the lower end of the range as of the last period of the data. We have cited anecdotal evidence that central bankers may "engineer" declines in the price of gold in financial markets, making speculation on the long side potentially treacherous. 

Gold as Insurance

If you're of a mind to purchase gold, you're probably in very good company. However, while gold purchases may be seen as a capital expenditure by the central banker, they are better viewed as insurance by the rest of us. Regular incremental, fixed-amount purchases thus become the payment of a "premium" and declining prices a windfall, because more "coverage" is gained and the cost basis is dollar-cost-averaged. Market timing ceases to be the overriding concern and good financial discipline is made less treacherous. It's good enough for central bankers, it ought to be good enough for the rest of us.

[1] Jerome F. Smith, Understanding Runaway Inflation, 1979 Edition, pp 48.