Tag Archives: monetary economics

Why QE didn’t send gold up to $20,000

This blog post is a guest post on BullionStar's Blog by the renowned blogger JP Koning who will be writing about monetary economics, central banking and gold. BullionStar does not endorse or oppose the opinions presented but encourage a healthy debate. 

Why didn't quantitative easing, which created trillions of dollars of new money, lead to a massive spike in the gold price?

The Quantity Theory of Money

The intuition that an increase in the money supply should lead to a rise in prices, including the price of gold, comes from a very old theory of money—the quantity theory of money—going back to at least the philosopher David Hume. Hume asked his readers to imagine a situation in which everyone in Great Britain suddenly had "five pounds slipt into his pocket in one night." Hume reasoned that this sudden increase in the money supply would "only serve to increase the prices of every thing, without any farther consequence."

Another way to think about the quantity theory is by reference to the famous equation of exchange, or

  • MV = PY
  • money supply x velocity of money over a period of time = price level x goods & services produced over that period

A traditional quantity theorist usually assumes that velocity, the average frequency that a banknote or deposit changes hands, is quite stable. So when M—the money supply— increases, a hot potato effect emerges. Anxious to rid themselves of their extra money balances M, people race to the stores to buy Y, goods and services, that they otherwise couldn't have afforded, quickly emptying the shelves. Retailers take these hot potatoes and in turn spend them at their wholesalers in order to restock. But as time passes, business people adjust by ratcheting up their prices so that the final outcome is a permanent increase in P.

In August 2008, before the worst of the credit crisis had broken out, the U.S Federal Reserve had $847 billion in money outstanding, or what is referred to as "monetary base"—the combination of banknotes in circulation and deposits held at the central bank. Then three successive rounds of quantitative easing were rolled out: QE1, QE2, and QE3. Six years later, monetary base finally peaked at $4.1 trillion (see chart below). QE in Europe, Japan, and the UK led to equal, if not more impressive, increases in the domestic money supply.

U.S. monetary base (banknotes and deposits at the Fed)

So again our question: if M increased so spectacularly, why not P and the price of gold along with it? Those with long memories will recall that while gold rose from $1000 to $2000 during the first two legs of QE, it collapsed back down to $1000 during the last round. That's not the performance one would expect of an asset that is commonly viewed as a hedge against excess monetary printing.

How Regular Monetary Policy Works

My claim is that even though central banks created huge amounts of monetary base via QE, the majority of this base money didn't have sufficient monetary punch to qualify it for entry into the left side of the equation of exchange, and therefore it had no effect on the price level. Put differently, QE suffered from monetary impotence.

Let's consider what makes money special. Most of the jump in base money during QE was due to a rise in deposits held at the central bank, in the U.S.'s case deposits at the Federal Reserve. These deposits are identical to other short-term forms of government debt like treasury bills except for the fact that they provide monetary services, specifically as a medium for clearing & settling payments between banks. Central banks keep the supply of deposits—and thus the quantity of monetary services available to banks—scarce.

Regular monetary policy involves shifting the supply of central bank deposits in order to hit an inflation target. When a central bank wants to loosen policy i.e. increase inflation, it engages in open market purchases. This entails buying treasury bills from banks and crediting these banks for the purchase with newly-created central bank deposits. This shot of new deposits temporarily pushes the banking system out of equilibrium: it now has more monetary services than it had previously budgeted for.

To restore equilibrium, a hot potato effect is set off. A bank that has more monetary services then it desires will try to get rid of excess bank deposits by spending them on things like bonds, stocks, or gold. But these deposits can only be passed on to other banks that themselves already have sufficient monetary services. To convince these other banks to accept deposits, the first bank will have to sell them at a slightly lower price. Put differently, it will have to pay the other banks a higher price for bonds, stocks, or gold. And these buyers will in turn only be able to offload unwanted monetary services by also marking down the value, or purchasing power, of deposits. The hot potato process only comes to a halt when deposits have lost enough purchasing power, or the price level has risen high enough, that the banking system is once again happy with the levels of deposits that the central bank has injected into the system.

What I've just described is regular monetary policy. In this scenario, open market operations are still potent.  But what happens when they lose their potency?

Monetary Impotence: Death of the Hot Potato Effect

A central bank can stoke inflation by engaging in subsequent rounds of open market purchases, but at some point impotence will set in and additional purchases will have no effect on prices. When a large enough quantity of deposits has been created, the market will no longer place any value on the additional monetary services that these deposits provide. Monetary services will have become a free good, say like air—useful but without monetary value. Deposits, which up to that point were unique thanks to their valuable monetary properties, have become identical to treasury bills. Open market operations now consists of little more than a swap of one identical t-bill for another.

Zimbabwe 100 trillion dollar notes together with gold bullion
100 Trillion Dollar Notes are not yet required to purchase gold. Why hasn't the increased money supply significantly increased the gold price?

When this happens, subsequent open market purchases are no longer capable of pushing the banking system out of equilibrium. After all, monetary services have become a free good. There is no way that banks can have too much of them. Since an increase in the supply of deposits no longer has any effect on bank behavior, the hot potato effect can't get going—and thus open market purchases have no effect on the price level, or on gold.

This "monetary impotence" is what seems to have overtaken the various rounds of QE. While the initial increase in deposits no doubt had some effect on prices, monetary services quickly became a free good. After that point, the banking system accepted each round of newly-created deposits with a yawn rather than trying to desperately pass them off, hot potato-like.

And that's why gold didn't rise to $20,000 through successive rounds of QE. Gold does well when people find that they have too much money in their wallets or accounts, but QE failed to create the requisite "too much money".

Should we Restore the Gold Standard?

This blog post is a guest post on BullionStar's Blog by the renowned blogger JP Koning who will be writing about monetary economics, central banking and gold. BullionStar does not endorse or oppose the opinions presented but encourage a healthy debate. 

Would it make sense to rebuild an international gold standard like the one we had in the late 1800s? Larry White says the idea has merit, David Glasner believes it isn't worth the risk. Over the years I've followed the back-and-forth between these two blogging economists, each of whom has done an admirable job defending their respective side for and against the gold standard. Let's look at one or two of the most important themes running through the White v Glasner debate.

Like a ruler measures distances, a nation's monetary standard serves as a measuring stick for the value of goods and services. People need to be able to set sticker prices with the unit, calculate profit and loss, negotiate labour contracts, and establish the terms of long-term debts using it. If the measuring stick is faulty, then all these important tasks becomes unnecessarily difficult.

Gold as Unit of Account

Since 1971 our measuring stick has been irredeemable paper currency, or a fiat money standard. Central banks try to ensure that, within the confines of their nation, the general level of domestic consumer prices stays constant, or at least rises at a constant rate of around 2-3%. And while the first decade of the fiat standard was a disaster characterized by high and rising inflation, central bankers in developed nations have generally managed to keep inflation on track for the last thirty or so years.

To re-establish gold as the measuring stick, each nation's unit of account—say the $ or ¥ or £—would have to be redefined as a certain fixed number of ounces of gold. Banknotes and central bank deposits, which are currently inconvertible, would be made convertible into an appropriate amount of gold. It is important that all nations return to the gold standard rather than just one, because one of the big advantages of an international gold standard is that with all currencies pegged to gold, it is much simpler for citizens of one nation to make calculations using another nation's unit. And this makes cross-border trade and investment easier to engage in.

Should banknotes and electronic fiat currency once again be made convertible into gold?

In Favour of the Gold Standard: Larry White

How well have the two standards served as measuring sticks? As the chart below illustrates, year-to-year changes in U.S. consumer prices were quite variable during the classical gold standard era, rising some years and falling the next. The source for this chart is from this paper that White has coauthored with George Selgin and William Lastrapes. The classical gold standard from which the authors draws their data lasted from 1880—when the majority of the world's major nations defined their currency in terms of the gold—to 1914 when the gold standard was dismantled on the eve of World War I. Data shows that the fiat standard that has been in place since 1971 demonstrates more predictable year-to-year price changes. Citizens of developed nations are pretty safe assuming that next year, domestic prices will rise by 2-3%.

Quarterly US inflation rate, 1875 to 2010

However, it is over longer periods of time that gold outperforms as a measuring stick. In the chart below, the authors show that the quarterly price level during the gold standard tended to deviate much less from its six-year average rate than during the fiat era. Because the general level of prices was more predictable under a gold standard, this provided those who needed to construct long-term debt contracts with a degree of certainty about where prices might be in ten or twenty years that is lacking under a fiat standard. White points out that this may be why 100-year bonds were common in the 1800s, but not so much now.

6-year rolling standard deviations of the U.S. quarterly price level

According to White, the main reason for the long-term stability of gold is the tendency for higher prices to encourage gold miners to increase the supply of metal, thus tamping down on the price, and conversely lower prices to encourage them to reduce production, thus buoying prices. In other words, prices under a gold standard were mean reverting. This mean reversion was generated "impersonally", or automatically, by the market, a superior sort of stability compared to that generated by a fiat standard, which depends on the skills and wherewithal of technocrats employed by the central bank.

Against the Gold Standard: David Glasner

David Glasner is skeptical about the gold standard because he doesn't agree that it mean-reverts fast enough. All of the gold ounces that have ever been mined continue to exist in vaults or under mattresses or around necks. Compared to this extant gold stock, the flow of new gold production is tiny. So if there is an increase in people's demand for gold, it is unlikely that new flows will be able to satisfy it, at least not for some period of time. Likewise, reduced gold production on the part of gold miners won't be able to vacuum up enough of the slack should people suddenly want less of the stuff. In either case, the price of gold will have accommodate shifts in demand by rising or falling quite a bit.

One thing that most monetary economists agree on is that fluctuations in the value of the item used as the standard—gold or fiat money—should not interfere with the "real" economy, say by causing unemployment or gluts of unsold goods. While many prices in an economy are incredibly flexible, like the price of stocks or gold or bitcoin, there are also many prices that are sticky, in particular labour. Under a gold standard, if there is a sudden increase in the demand to hoard gold, then there will be pressure on price of gold to rise. The rise in the gold price means that the general level of prices must fall. Goods and services, after all, are priced in terms of gold-backed notes. But with wages and many other prices locked in place, the response on the part of employers will be to adjust by announcing mass layoffs. Rather than cutting the sticker prices of goods, retailers will suffer though gluts of unsold inventory. This is a recession.

Glasner's favorite example of this occurred during the late 1920s. After WWI had ended, most nations attempted to restore the pre-war gold standard with banknotes once again being redeemable with fixed amounts of gold. But then the Bank of France, France's central bank, began to buy up huge quantities of gold in 1926, driving the gold price up. The U.S. Federal Reserve was unwilling to counterbalance what was viewed as insane purchases by the Bank of France, the result being the worst recession on record, the Great Depression.

What Type of Gold Standard?

Given that various commodity standards have been in place for centuries, why did it take till 1929 for a massive monetary mistake to finally occur? White blames this on large government actors, specifically central banks. In the initial international gold standard that ran from 1880-1914, nations such as Canada, Australia, and the U.S. didn't have central banks. Commercial banks in these nations chose to link their privately-issued banknotes to gold, the goal of these competing banks being to to earn profit rather than enact social policies. So earlier versions of the gold standard functioned far more naturally, without the meddling of large actors who refused to abide by the typical rules of a gold standard. It is for this reasons that White prefers that any return to the gold standard be packaged with an end to central banks, thus precluding episodes like the Great Depression from occurring.

David Glasner remains skeptical. According to Glasner, even the classical gold standard that ran from 1880 to 1914 required management, the Bank of England leaning in such a way as to counterbalance large demands for gold from other central banks and thus preventing anything like the Great Depression from occurring. And even if central banks were to be dismantled under a 21st century version of the gold standard so as to preclude an "insane" Bank of France scenario, there remains the problem of "panic buying" of gold by the public—and the resulting gold-driven recession this would cause.

So Where does that Leave us?

As I hope you can see by a quick exploration of the debate between Larry White and David Glasner, restoration of the gold standard is a complicated issue. I'd encourage readers who are interested to dive a bit deeper into the subject by reading David's posts here and Larry's here.

As for myself, White's work on the 1880-1914 gold standard has been helpful in removing many of the preconceptions I had of the gold standard, no doubt passed off to me by commentators who were never very familiar with the actual data. Nevertheless, I tend to agree with Glasner that under a global gold standard (with no central banks) a sudden spike in the public's demand for gold would impose large costs on the global economy. With citizens of the globe being so connected through the internet and free capital markets, these sorts of episodes might be more common nowadays than they were in the 1800s. I'm not sure the benefits of a gold standard, including exchange rate stability, make up for this risk. Given that Western central banks have done a fairly decent job of keeping inflation under control for the last thirty or so years, I'll give them the benefit of the doubt... for now.