« What I've Been Reading | Main | Robbery fact of the day »
Should we consider a gold standard?
Lawrence H. White responds to critics. In my Cato email I received:
"A gold standard does not guarantee perfect steadiness in the growth of the money supply, but historical comparison shows that it has provided more moderate and steadier money growth in practice than the present-day alternative, politically empowering a central banking committee to determine growth in the stock of fiat money," White concludes. "From the perspective of limiting money growth appropriately, the gold standard is far from a crazy idea.”
"Far from a crazy idea," OK. But would you push the button for it? I say no. There is little doubt that over the broad sweep of world history, commodity standards have outperformed paper money. But we don't live in the broad sweep of world history, we live in 2008 and our ability to monitor and control central banks is unparalleled. The central banks of the wealthier nations work pretty well. My main worry with the gold standard is simply the pro-cyclicality of the money supply and for all its talk of money demand the paper doesn't much address this concern. For instance would you really want a contracting money supply in today's environment? And yes credit crunches of this kind happen in market settings too so you can't blame it all on Alan Greenspan.
And I am not reassured by this (admittedly true) sentence: "At the right reentry rate, dollar prices would not need to jump [from the transition]." One five or ten percent deflation is enough to crush the economy and indeed the whole gold standard idea. Given the socialist calculation debate, can we really know the right transition price? Gold at $900 an ounce? $600 an ounce? Anybody?
Addendum: Here is White's response; see also the exchange in the comments with White and others.
Posted by Tyler Cowen on February 9, 2008 at 07:13 AM in Economics | Permalink
Comments
The purchasing power of gold has varied over a factor of ten since the 70s. Hardly a standard of value. More like pork belly futures.
Posted by: Ron Hardin at Feb 9, 2008 7:43:00 AM
Given that (unstable, massively pro-cyclical) non-monetary demand for gold swamps (more stable, less cyclical) monetary demand for gold, short of getting the rest of the world on the gold standard first, a gold standard is going to be trouble. Most of the gold standard's history in which it performed so well was when much of the world was on the standard and so monetary demand greatly exceeded non-monetary demand.
If you're looking for a return to commodity-backed money, gold seems a singularly silly commodity to choose.
Posted by: Grant Gould at Feb 9, 2008 8:19:10 AM
the price of gold would have to be set much higher then any foreseeable market clearing price. This was one of the problems in the 1960s when the gold price was too low and private speculators kept buying gold from governments in anticipation of a higher price.
At stable gold prices, industrial, jewelery and third world gold demand grows faster than both world real GDP and the supply of gold.
As a consequence over the long run the real price of gold has to keep rising.
If gold is the base for the money supply this implies that all other prices must fall -- deflation.
Gold based money systems worked when we measured economic growth in terms of decades or centuries. But if we are going to experience rapidly rising world wide standards of living the gold supply is inadequate.
Posted by: spencer at Feb 9, 2008 8:50:39 AM
Why not just end legal tender laws and let the market decide how or whether to transition out of fiat money?
Posted by: Selfreferencing at Feb 9, 2008 8:59:58 AM
Anybody have references to the purchasing power of gold over the last few decades? I couldn't find any good info through googling.
Posted by: ben abuya at Feb 9, 2008 9:22:08 AM
P.S. If the real price of gold had grown at a 3% annual rate from $39 in 1969 it would now be %833.
The $39 price in 1969 probably was a market clearing price as their were no significant central bank sales or purchases of gold in 1969.
Posted by: spencer at Feb 9, 2008 9:28:41 AM
Sure, we can consider a gold standard--let's consider how many times its failed to be enforced. Gold has historically had many issues over the past couple of centuries--when it flees a nation-state, it causes bank failures and a return to fiat money. Today governments have accumulated so much gold in their reserves (think of Roosevelt's 1933 gold seizure and the law preventing its use as legal tender) from that they can effectively inflate its value any time they want if they don't redefine it by fiat: governments cannot prevent themselves from redefining its value for monetary interventions, its like saying that governments can reduce spending or lower taxes in the long term.
Today you cannot simply "go back" to a gold standard unless you also reign in debt-based money issued by central, commercial and personal banking systems. Governments especially appreciate what bankers can do for them when they want to fund new wars. "Gold standard? What? There's a war to be fought! There's money to be made!" Don't believe this? Gold has failed many, many times. Just in America alone we have: 1797 (gold flees Britain to France after the French institute a gold standard to solve their hyperinflation--gold starts leaving New York causing bank failures), 1819 (contraction after the War of 1812, funded by government borrowing), 1837 (banks stopped paying in specie, after the western real estate bubble where people borrowed money from banks to buy public lands at $1.25/acre), 1857 (discovery of gold in California causes inflation, the collapse of speculative investments in railroads, followed by public confidence disruptions like the loss of 30,000 pounds of gold at sea off the coast of North Carolina in a hurricane), 1873 (speculative investments in railroads collapse, again more bank paper issued in unrestrained credit--makes you think of today's Internet bubbles, eh?). It keeps going: 1884 (gold reserves in Europe are depleted, US Treasury halts investements and calls in loans), 1893 (public confidence in banks falters and the banks can no longer redeem notes for gold, but still for silver), 1896 (silver reserves are depleting), and 1907 (banker's panic caused by retraction of loans).
Today I think we're beginning to see the great unwinding of debt-based monetary expansion that has lasted almost 100-years since the creation of the Federal Reserve and the last 30 years or so since we left the Bretton Woods system to a purely fiat currency. Going back to a system that manages the supply of currency to an economy isn't a crazy idea, its the getting of governments and bankers to stick to it--that's the crazy part. They never do.
Posted by: Simon Janes at Feb 9, 2008 10:13:34 AM
If you were to go to a gold standard, it would make little sense to have it linked to a nominal amount of dollars at a fixed rate. I think it would make more sense to have bills that represent say 0.01 or 0.5 grams of gold and bank accounts denominated in grams (or moles) of gold. This would mean no more dollars ($) being used at all. Not a very likely scenario, but I think that it would be the best, most practical way to use gold as a currency.
Posted by: gld at Feb 9, 2008 10:13:40 AM
This is about the same as trying to decide what the best style of dress is -- clearly there is no objectively "best" type, just that which people tend to prefer. The debate should be: should the government (the Constitution notwithstanding) decide and manage for us what we are to use as a medium of exchange? I say no. As with clothing, let the market decide. It's quite the opposite of revolutionary.
Posted by: Jeremy Sapienza at Feb 9, 2008 10:29:36 AM
Joseph Schumpeter on the larger significance of the gold standard--both for and against:
“An ‘automatic’ gold currency is...is extremely sensitive to government expenditure and even to attitudes or policies that do not involve expenditure directly, for example, to foreign policy, to certain policies of taxation, and, in general, to precisely all those policies that violate the principles of liberalism. This is the reason why gold is so unpopular now and also why it was so popular in a bourgeois era. It imposes restrictions upon governments or bureaucracies that are much more powerful than is parliamentary criticism. It is both the badge and the guarantee of bourgeois freedom—-of freedom not simply of the bourgeois interest, but of freedom in the bourgeois sense. From this standpoint a man may quite rationally fight for it, even if fully convinced of the validity of all that has ever been urged against it on economic grounds. From the standpoint of etatisme and planning, a man may not less rationally condemn it, even if fully convinced of the validity of all that has ever been urged for it on economic grounds."
(Schumpeter, 1954, 405-406.His italics.).
That quotation concludes Richard Timberlake's critique of Eichengreen:
http://www.econjournalwatch.org/main/intermedia.php?filename=TimberlakeIntellectualTyrannyAugust2005.pdf
Posted by: Daniel Klein at Feb 9, 2008 10:40:11 AM
My main worry with the gold standard is simply the pro-cyclicality of the money supply and for all its talk of money demand the paper doesn't much address this concern. For instance would you really want a contracting money supply in today's environment? And yes credit crunches of this kind happen in market settings too so you can't blame it all on Alan Greenspan.
The money supply under the gold standard would be less cyclical than under a fiat standard. You are wary of a contracting money supply in today's environment, but ignore the expanding money supply caused by Easy Al hitting the monetary accelerator/panic button on numerous occasions. In a market economy, the evidence is that credit crunches would be smaller and the fallout from them less painful.
Greenspan should be in leg irons and should have to disgorge the ill-gotten profits from his book.
Posted by: Bill Stepp at Feb 9, 2008 10:41:27 AM
All that is really required is that the right to own, buy, sell or lend
gold be guaranteed by constitutional amendment. To thrwart the
connivances of lawyers we might include language guaranteeing the right
to trade or barter gold for anything, and trade or barter anything for
gold. This would enable people through the market to establish whatever
degree of gold standardness they want.
As for the contraction of money supply as a consequnce gold standard.
This is difficult to fathom. Gold is nearly indestructible. So if
anything the supply of gold of money expands rapidly in an economic
contraction. Remember, the supply of money is relative to the amount
goods and services that can be purchased for a given unit. A deflation
represents an automatic expansion of money supply. An inflation is
an automatic contraction of money supply. That is why central banks
cannot cut rates fast enough in an inflationary recession.
Posted by: joebeck at Feb 9, 2008 10:48:39 AM
Who audits the gold reserves? Even with a gold standard, governments have ample room for shenanigans. And why gold, other than for historical reasons? Why not platinum or some other substance?
Posted by: at Feb 9, 2008 10:51:37 AM
joebeck -- you are considering the case of a closed economy where there can be no flow of gold into or out of a country. But an individual country can experience a gold outflow -- and it happened regularly when the world was on a gold standard.
Posted by: spencer at Feb 9, 2008 10:56:24 AM
Free market money is the proper way to go. Combined with the monopoly on coercion, a government controlled currency (even with nominal commodity backing) is a recipe for sure disaster. Legal tender laws, along with capital gains taxes on non-dollar transactions should be eliminated.
I must say I am quite stunned by Tyler's faith in today's central bankers.
I don't expect free market money to return until the present debt-as-currency regime collapses- but collapse it will.
Posted by: Yancey Ward at Feb 9, 2008 10:56:53 AM
Selfreferencing and Jeremy - it's a myth that you can't start your own currency. In fact, you can. You just have to tell the IRS about it and be sure and pay taxes regardless of what currency you're using. Check out the Ithaca Hour.
So no, we aren't all using dollars because we have no choice. You can feel free start your own gold-backed currency and watch as the world doesn't scramble to your door.
Posted by: Jon Kay at Feb 9, 2008 12:07:24 PM
"But we don't live in the broad sweep of world history, we live in 2008 and our ability to monitor and control central banks is unparalleled."
Why doesn't the Fed publish the data on M3 anymore?
Posted by: Pippin at Feb 9, 2008 12:13:44 PM
John Kay,
Tell that to the makers of the Ron Paul Dollar. The feds scrambled to their door pretty quickly.
http://www.thestreet.com/markets/marketfeatures/10390631.html
Posted by: Sam at Feb 9, 2008 2:13:07 PM
E-Gold's been having some trouble, too: E-Gold.
Posted by: ben abuya at Feb 9, 2008 2:21:05 PM
"But we don't live in the broad sweep of world history,..."
I love experimentation. In fact, I pay a high personal cost to be on the leading edge of gadgetry (and political thought).
But, the odds are highly in favor of things working in the past still working.
Just because something worked in the past doesn't mean it will work now. But just because they worked in the past doesn't mean they won't work now.
I'd push that button. I'd keep pushing it. I'd get some duct tape and tape that button down. I'd put a box over the button and write "tarantula" on the box. I'd put a "Caution, wet floor" sign in front of the button.
Posted by: Andrew at Feb 9, 2008 2:34:45 PM
I reply to Tyler's post here .
Many of the anti-gold-standard arguments in the previous comments on Tyler's post are addressed in my Cato Briefing Paper, available at the link Tyler provides.
Posted by: Lawrence H. White at Feb 9, 2008 2:47:52 PM
As I describe here, the "right transition price" can be computed trivially by:
(1) calculating the number of dollars in the world.
(2) calculating the amount of gold backing those dollars.
(3) dividing (2) by (1).
The trouble, as I discuss, is that this yields a number at least 10x, and maybe 100x, higher than the present price.
Why so vague? Because measuring the number of dollars in the world is almost impossible. This is problem A. Problem B is how you actually enact a policy whose consequence is an instant order-of-magnitude gap in the price of a widely traded monetary commodity. Problem B is almost impossible as well. That's life in the big city.
Professor White (a quasi-Austrian of the "free banking" school) gets around both problems easily, because what he is proposing is not a monetary system in which gold is money, but a monetary system in which currency is "backed" by gold. Therefore, the quantity of gold in the world is not a mathematical limit on the quantity of money.
Mainstream Austrians believe that these fractional gold standards (the 19C Bank of England "classical" gold standard, the post-WW1 "gold exchange standard," and the ultimate pseudo-gold-standard, Bretton Woods) served as a slippery slope on the road to paper money. They would prefer to return to a gold standard in which money is simply defined by weight of gold, and under which banks have no way to expand that supply.
Professor White believes that it's possible to restore a Bagehotian fractional-reserve gold standard, in which paper notes are redeemable for gold on demand, but there is not sufficient gold in the vault to redeem them all at once. More broadly, he supports (unlike mainstream Austrians) a banking system in which financial intermediaries (such as banks) systematically transform maturities, leading to a systemic mismatch between assets and liabilities of a given maturity date.
Given that we are currently in the midst of a massive liquidity crunch caused by just this practice, I think the good professor would do well to expand on his assertion that it is benign, in the absence of a lender of last resort with an effectively infinite money supply (the Fed's printing press). This explanation
“Fiat money is necessary so that a lender of last resort can meet the liquidity needs of the banking system.” History shows that a lender of last resort would hardly be needed, given a stable monetary regime and a sound banking system (again it is instructive to contrast the United States with Canada in the 19th century). In the rare cases such a lender might be needed, bank clearinghouses could play the role.
certainly leaves me wanting more, and strikes me as suspiciously empirical rather than praxeological. To an Austrian, the phrase "history shows" generally translates as "I choose to assume."
Historical experience is especially questionable on this point for two reasons.
First, a high-friction financial environment, as found in the past, is conducive to the maintenance of unnatural pegs - such as a peg between gold and paper banknotes with a high paper-to-gold ratio. In a modern, frictionless electronic market, once paper trades down to 0.9999% of gold, arbitrage will take over and the bank run is on. Currency pegs do not bend in the 21st century. They shatter.
Second, the relationship between the State and banking is traditionally close and traditionally informal. How "free" was the "free banking" era, really? The Bank of England was a nominally private corporation well into the 20th century, but it was a parastatal from day one. It is very difficult for historians to assess the informal political connections between formally "free" banks and the governments who, in theory, should be treating them as independent. Sweetheart treatment in bankruptcy cases, acceptance of banknotes as tax payments, etc, etc, etc, are all very convincing motivations for customers to patronize unsound banks. And the difference between protecting an insolvent bank (the classic example being the Bank Restriction of 1797) and irredeemable paper currency is, as history shows, a difference of degree.
Posted by: Mencius at Feb 9, 2008 3:59:01 PM
Tell that to the makers of the Ron Paul Dollar. The feds scrambled to their door pretty quickly.
I suspect there's more to that than meets the eye. The claim in the article is that users were trying to "spend" the RP dollars. What does that mean? It's obviously legal to barter medallions with Ron Paul's face or any other design on them for some other goods or services. I myself recall bartering pieces of cardboard with baseball players' faces on them for other pieces of cardboard with other players' faces. I never got arrested for this, nor did my friends who did the same, despite our activities being open and widely known.
OTOH, it's surely illegal to claim that the medallions are "dollars" in a meaningful sense. Are you sure you know what these guys are accused of?
Posted by: Bernard Yomtov at Feb 9, 2008 4:21:35 PM
Lawrence White -- your comment that gold is counter-cyclical is inconsistent. As you observe the physical supply is counter-cyclical, but this does not address the question of what happens to the nominal value of the gold supply, and that is what is important to the money supply. Your argument seems to switch back and forth between a world where the price of gold is fixed and one where the price is free to fluctuate. If the price is fixed, then the supply is not counter -cyclical as you argue. If the price is free you offer no evidence that the drop in supply will cause the total nominal dollar value of the supply of gold to contract as your theory requires. This theory requires some strong assumptions that demand elasticity is also weak enough to offset the negative supply elasticity. Otherwise, the drop in physical supply will cause the nominal value of the gold supply to act in a pro-cyclical basis as Tyler fears rather then in a counter-cyclical basis as you argue.
For those not familiar with the gold market the supply argument centers on South Africa.
The dominant supplier of gold is South Africa and the price elasticity of South African gold is negative. Yes, higher prices cause the supply of South African supply to fall in the short to medium run.
Consequently, on a cyclical basis when the price of gold rise the supply contracts, not expands as standard theory says.
South African supply is negative because South African mine companies are not trying to maximize short run profits. Rather they work on a principle of adequate short run profits to cover the firms dividend while in the long run they are trying to maximize the life of the mine. South African gold mine are very deep mines -- the gold is several miles below the surface. Consequently, the bottleneck in gold production is the capacity of the mines to haul the unprocessed ore up the mine shaft. They actually lift several tons of raw ore for each ounce of gold. They make the quality of the ore they mine an inverse function of the price of gold. As the price of gold rises they shift production to the lowest quality veins that can be mined at that price. This leaves the richer veins to be mined at some future date when prices or lower or expenses are higher. This is how they maximize the long run life of the mine and the amount of gold that can be obtained from a given mine or gold vein. This process makes the short to intermediate run -- what matters on a cyclical basis -- supply elasticity of gold negative. So on a short run basis when a strong economy causes the price of gold to rise the supply contracts, causing the price to rise even more. Of course the reverse happens in a recessionary environment when the gold price tends to fall the supply will expand causing the price to contract even further.
Posted by: spencer at Feb 9, 2008 4:25:23 PM
lawrence White --- you keep referring to the case of Canada. But Canada is a special case because they only have a few national banks. Their system is an oligopoly where the central banks plays a very different role then in a competitive banking system. This is why the history of the Canadian system, especially in the depression has little relevance to the argument you are making.
Posted by: spencer at Feb 9, 2008 4:33:30 PM
