December 12, 2005
The gold standard and the Great Depression
How the gold standard contributed to the Great Depression.
There always seem to be voices raising the possibility that a return to a monetary gold standard could solve all our problems. Among those championing this meme this week were Chris Mayer at Daily Reckoning, Robert Blumen at Mises Economics Blog, and some of my fellow blogjammers.
Under a pure gold standard, the government would stand ready to trade dollars for gold at a fixed rate. Under such a monetary rule, it seems the dollar is "as good as gold."
Except that it really isn't-- the dollar is only as good as the government's credibility to stick with the standard. If a government can go on a gold standard, it can go off, and historically countries have done exactly that all the time. The fact that speculators know this means that any currency adhering to a gold standard (or, in more modern times, a fixed exchange rate) may be subject to a speculative attack.
After suspending gold convertibility in World War I, many countries stayed off gold and experienced chaotic fiscal and monetary policies in the early 1920's. Many observers reasoned then, just as many observers reason today, that the only way to restore fiscal and monetary responsibility would be to go back on gold, and by the end of the 1920's, most countries had returned to the gold standard.
I argued in a paper titled, "The Role of the International Gold Standard in Propagating the Great Depression," published in Contemporary Policy Issues in 1988, that counting on a gold standard to enforce monetary and fiscal discipline in an environment in which speculators had great doubts about governments' ability to adhere to that discipline was a recipe for disaster. International capital flows became more erratic, not less, as doubts were raised about whether first the pound would be devalued and then the dollar. Britain gave in to the speculative attacks and abandoned gold in 1931, whereas the U.S. toughed it out by deliberately raising interest rates in 1931 at a time when the economy was already near free fall.
Because of this uncertainty, there was a big increase in demand for gold, the one safe asset in this setting, which meant the relative price of gold must rise. If everybody is trying to hoard more gold, you're going to have to pay more potatoes to get an ounce of gold. Since the U.S. insisted on holding the dollar price of gold fixed, this meant that the dollar price of potatoes had to fall. The longer a country stayed on the gold standard, the more overall deflation it experienced. Many of us are persuaded that this deflation greatly added to the economic difficulties of those countries that insisted on sticking with a fixed value of their currency in terms of gold.
Ben Bernanke and Harold James, in a paper called "The Gold Standard, Deflation, and Financial Crisis in the Great Depression: An International Comparison" published in 1991 (NBER working paper version here), noted that 13 other countries besides the U.K. had decided to abandon their currencies' gold parity in 1931. Bernanke and James' data for the average growth rate of industrial production for these countries (plotted in the top panel above) was positive in every year from 1932 on. Countries that stayed on gold, by contrast, experienced an average output decline of 15% in 1932. The U.S. abandoned gold in 1933, after which its dramatic recovery immediately began. The same happened after Italy dropped the gold standard in 1934, and for Belgium when it went off in 1935. On the other hand, the three countries that stuck with gold through 1936 (France, Netherlands, and Poland) saw a 6% drop in industrial production in 1935, while the rest of the world was experiencing solid growth.
A gold standard only works when everybody believes in the overall fiscal and monetary responsibility of the major world governments and the relative price of gold is fairly stable. And yet a lack of such faith was the precise reason the world returned to gold in the late 1920's and the reason many argue for a return to gold today. Saying you're on a gold standard does not suddenly make you credible. But it does set you up for some ferocious problems if people still doubt whether you've set your house in order.
Nevertheless, I'm willing to grant Tim Iacono that the stuff is pretty.
Posted by James Hamilton at December 12, 2005 12:58 PMdigg this | reddit
Listed below are links to weblogs that reference The gold standard and the Great Depression:
» Good as gold from Asymmetrical Information
James Hamilton has an excellent post on why the gold standard isn't the panacea everyone things: Under a pure gold standard, the government would stand ready to trade dollars for gold at a fixed rate. Under such a monetary rule, it seems the dollar is ... [Read More]
Tracked on December 12, 2005 02:18 PM
» Gold Standard: Bad Idea from Dean's World
Econbrowser explains why Gold Standards for currency aren't such a hot idea.
I'm also often amused by how often I see ads saying that in these...[Read More]
Tracked on December 13, 2005 05:05 AM
» Gold, inflation and smashing the state from BradSpangler.com
Kudos to Jason Ballot for handling some of my light work. No, really. I’ve been meaning to get around to posting in response to James Hamilton’s recent post, The gold standard and the Great Depression. Jason, however, beat me to the punch... [Read More]
Tracked on December 15, 2005 02:42 PM
» Jim Hamilton on the Gold Standard and the Great Depression from Brad DeLong's Website
Jim Hamilton writes on the gold standard and the Great Depression. He takes the standard modern-macro line, and he expresses it very well. If your government doesn't have monetary-policy credibility, attempting to establish that credibility by going on... [Read More]
Tracked on December 19, 2005 11:48 AM
So countries went off the gold standard at different times... over a number of years. How was the arbitrage pressure handled? While I realize they didn't have 'electronic exchanges' exactly like we do today... they were able to wire funds or in the very least wire 'letter of credit' to then borrow in the local currency & effect trades the same way.
I would guess the arbs - as now - would raise all kinds of mischief... did they? To what effect?
Posted by: dryfly at December 12, 2005 01:48 PM
I've never understood why pegging any given currency to any other was considered to be a good idea. Especially in today's world, you would think the lightning-fast arbitrage trades would bring the peg down very quickly.
Posted by: Ben Fulton at December 12, 2005 02:10 PM
Dryfly, if country A went off gold but countries B and C did not, the exchange rate of A vs. B and of A vs. C would both immediately change with the exchange rate of B vs. C staying where it was before. So, there's no arbitrage just because A goes off first, then later B, then later C. The arbitrage comes when you expect that B is going to go off next year but it hasn't yet done so today. This arbitrage is fierce-- it's the speculative attack I was talking about, and it's exactly what the problem is all about.
Posted by: JDH at December 12, 2005 02:16 PM
The current floating rate system works; if it works, don't fix it!
Posted by: Jack Miller at December 12, 2005 02:23 PM
I don't assume the state should issue fiat currency in the first place, as was the case before the central bank was created. Verification of assets was possible even then. The process was slow in the past, but the state was even slower. The (federal) state actually had to force citizens to adopt the fiat currency via laws regarding acceptable payments.
Posted by: Siggyboss at December 12, 2005 02:45 PM
Can any exchange regime - fixed or floating - survive an episode of gross fiscal irresponsibility, especially when the Central Bank decides to monetarize the fiscal recklessness?
Posted by: pgl at December 12, 2005 02:52 PM
Jack Miller -- "The current floating rate system works; if it works, don't fix it!"
Are you serious?
Posted by: Movie Guy at December 12, 2005 03:20 PM
The arbitrage comes when you expect that B is going to go off next year but it hasn't yet done so today. This arbitrage is fierce-- it's the speculative attack I was talking about, and it's exactly what the problem is all about.
Did they have those kinds of 'attacks' in the 30s? I've not read about them. Heard a lot about the 'Crash' and 'Smoot' and the 'Great Inflation' prior to the 'Crash'.
Plus I've read a little about the arbitrage attacks resulting in the 1997 'asian contagion' and in the 70s when BW failed.
But know nothing about the history surrounding arbitrage in the 30s.
Just curious if there is a readable economics history book you could recommend (not too academic for my poor ignorant head) covering the 30s on this particular aspect. I have family members who would also be quite interested...
Anyone else? TIA
Posted by: dryfly at December 12, 2005 03:34 PM
Setting aside the current bull gold market, how about the Nasdaq or Russell 2000 since the lows in year 2002?
Posted by: nate at December 12, 2005 04:31 PM
Over the Thanksgiving weekend I lost a little more respect for fiat currency - I spoke with someone who was unloading cargo at the Baghdad Airport during the summer of 2004. He was there to help unload a C5 containing twelve pallets of $100 bills, just before the handover from the CPA to the interim Iraqi government. Each pallet had many shrink wrapped bundles of newly minted bills, and all the bundles were shrink wrapped together. Each pallet totalled a little over $200 million for a grand total of about $2.5 billion (this drop is well-documented - just search on "$2.5 billion Baghdad").
The plane landed in the middle of the night and taxied up next to two Chinook helicopters and four deuce-and-a-halfs. They were quickly encircled by 20 military vehicles and about a hundred soldiers. Two pallets were loaded directly into each helicopter and they were gone within minutes. The remaining pallets had to be broken down and the smaller bundles were thrown into the back of the four trucks. This took about 25 minutes and the four trucks then left the airport, escorted by the other 20 vehicles and the hundred soldiers. This drop was unusual because of its size - the regular load is two pallets, $400 million, on a C-130.
Posted by: Tim at December 12, 2005 05:23 PM
"A gold standard only works when everybody believes in the overall fiscal and monetary responsibility of the major world governments and the relative price of gold is fairly stable."
Oh. And a fiat standard just completely works without qualification. (laff, laff, laff)
Posted by: Billy Beck at December 12, 2005 07:41 PM
Billy scores a good point here.
So let's see... the government screws around with fiscal policy, creating uncertainty all over the place, so investors flee to the one thing that isn't subject to a government's whim -- a physical value, gold -- and it's the gold standard that's at fault? And the solution to all that certainty, is to give even more discretionary power to the government?
Gold does not cause uncertainty; outside of nature, only government can create it at such scales.
Posted by: Seerak at December 12, 2005 09:17 PM
One thing to remember is that gold has value because of its scarcity.
What happens when gold becomes as common as, say, copper is now?
Not possible, you say? Think again-- one of these days (probably not anytime before 2050, but certainly by 2100) we'll be mining the moon and asteroids for metals, and there's a LOT of material out there to be found. Of course, our economies will be a lot bigger, too, but not big enough to justify a price of many hundreds of dollars an ounce for gold when the stuff will probably be available in the same quantities as copper is now.
There are outfits out there that would be willing (in an era of asteroid mining) to sign a futures contract for gold at, say, $100 an ounce if the other party can promise one-ton deliveries on a regular (once a month?) basis, payable upon delivery.
And the price will only fall farther as supply becomes more plentiful.
In the short term, maybe gold is a wise buy. But for the long term (with an eye to passing on assets to grandchildren), buy a well-diversified basket of stocks and bonds-- betting on the long-term prospects of American business is better than placing all of one's money in a depreciating asset like gold.
Posted by: Hale Adams at December 12, 2005 11:25 PM
"I've never understood why pegging any given currency to any other was considered to be a good idea. Especially in today's world, you would think the lightning-fast arbitrage trades would bring the peg down very quickly."
Even if this is true, floating exchange rates create uncertainty. If I am doing business in another country, I have to worry about exchange rate fluctuations in addition to all my other sources of risk. There are ways to deal with this (hedging with futures, etc...) but these all entail real costs. This may not justify fixed exchange rates (of which the gold standard is a form), but a floating system is not without its downsides.
Posted by: Anonymous at December 12, 2005 11:34 PM
Oh, and Tia - read "Golden Fetters" by Barry Eichengreen. Amazon link here.
Posted by: Anonymous at December 12, 2005 11:37 PM
The _real_ value of gold has varied over a factor of TEN since the 70s, not a good sign for a store of value.
The store of value seems to have been buying and rolling over 2y treasuries. They more or less match the interest rate without plummeting in value when the interest rate changes.
Posted by: Ron Hardin at December 13, 2005 12:50 AM
A gold standard only works when everybody believes in the overall fiscal and monetary responsibility of the major world governments and the relative price of gold is fairly stable.
Cart/horse inversion. The point here is that a gold standard would force fiscal and monetary responsibility upon a nation. Rather like starting a 12-step program. I agree, though, the US is on too big of a bender at the moment. And for the past 40-50 years.
The _real_ value of gold has varied over a factor of TEN since the 70s, not a good sign for a store of value.
Oh, puhhllleeze. I'd argue it's the value of the dollar that has vacillated by a factor of TEN, not gold. Early 1980s, an ounce of gold could buy the Dow Jones Industrials (quoted price). Now you need ~20 ounces of gold to buy the Dow. Does this mean gold has lost 95% of it's value? If so, then the stock market has gone nowhere. Or has gold kept its value and the market shot up by a factor of 20? You can't have it both ways.
Instead of comparing vs. paper assets, compare vs. real assets. An ounce of gold can buy a decent man's suit. Sometimes a better brand, sometimes a lesser brand, but always a decent suit that can be worn in middle-class company. Near as we can tell, it's been that way since gold has been used as money.
When the US Dollar goes the way of the 1923 German mark, gold will still have its purchasing power. The Roman Empire is dust, but you can still buy goods and services with its gold coins, even if you need to convert them to the local currency du era first.
The store of value seems to have been buying and rolling over 2y treasuries. They more or less match the interest rate without plummeting in value when the interest rate changes.
This is a fine approach as long as the dollar is stable and inflation is low. It will probably continue to work for some time. Internationalism is deflationary (good for bonds) and US bonds are liquid and trustworthy. It will not be that way forever! It took GM about 40 years to get to junk bond status; it should take the entire US somewhat longer.
Posted by: Anonymous at December 13, 2005 04:54 AM
There were speculative attacks long before the 1930's. A very nice book on the issue is Velde and Sargents' "The big problem of small change"
Posted by: Virgilio at December 13, 2005 04:55 AM
I was just telling a friend how my son's college tuition at Carnegie-Mellon was ten times my own college tuition in the late '60s, how houses cost ten times more,as does a car and a gallon of gas. The Dow Jones Industrial Average is roughly ten times what it was back then too. Air fares L.A. to SF are ten times more as is the cost of the NY-DC shuttle(more or less).
My son's starting salary and those of his friends were not ten times mine in 1970, so I guess that all can be defined as a drop in the standard of living during that time.
Posted by: jim miller at December 13, 2005 04:55 AM
>An ounce of gold can buy a decent man's suit.
But that is as much of a fiat decision as is anything involving money. Gold is valuable because we define it as valuable; tubuluar fullerenes longer than a centimeter are equally scarce, but nobody is calling for a standard based on that.
I think that most people who call for a gold standard have a mental image of going to the bank, getting a big bag of gold coins, then dumping them in a vat and swimming around like Scrooge McDuck.
Posted by: DensityDuck at December 13, 2005 05:25 AM
Jim has an interesting point on college tuition.
Anyone else have input on this?
Posted by: nate at December 13, 2005 06:18 AM
I use cars as my standard... in the 1950s you had to put 1/3 down and payoff in 3 years...who can afford that now??
10 percent down and lease and pay off in 8 years??
transportation then cost 3 years of your life and now costs 8..
Posted by: embutler at December 13, 2005 06:35 AM
Ah, things just aren't like they used to be. That car in the '50s didn't have air conditioning, power steering, power brakes, and turn signals were an option. In 1956 our family took its first trip to the Rockies. Every car had a canvas bag of water straped to the radiator so that water could be added when the car overheated, and you didn't drive far without seeing many cars stopped with their hood up. Tires lasted 15,000 to 20,000 miles.
The house that was so cheap was probably 1,000 square feet with one bathroom, no garage, yada yada.
In 1964 consumables comprised 43% of personal consumption expenditures compared to 29% last quarter. This is the cost of food, clothing and energy. The cost of durables (vehicles, appliances and furniture) totaled 14% of personal consumption expenditures in 1964 versus 12% now.
If you want to buy a vehicle for $12,000 it is both available, and better than most of the 1950's vehicles.
in 1967 I received a bonus of a month's salary. I used the entire month's salary to buy our first color TV. Want to compare that to my new High Def -
Get real - Enjoy life.
Posted by: Bill Ellis at December 13, 2005 06:53 AM
Gold is not money and money is not gold.
The idea of legally linking those two things is idiotic and nonsensical, and is based on superstition and a very primitive understanding of economics and money.
Everyone say it with me, I don't even have it trademarked:
GOLD IS NOT MONEY AND MONEY IS NOT GOLD
Posted by: Kevin F at December 13, 2005 08:04 AM
Americans own assets of $62 Trillion Dollars, owe $11 Trillion in debts to leave a net worth of $51 Trillion. This net worth increased by $4 Trillion Dollars in the past 4 quarters. Household savings is $4.7 Trillion and Consumer debt is $3 Trillion.
In the face of real economic progress, folks sit around talking about standards of living going down because college tuition has gone up. Do you think that maybe college tuition has gone up because millions more are willing to pay for this luxury?
Yes, I was serious when I said, "If it works, don't fix it." Compare the panics and depressions of past centuries to the relative calm of today's markets and you will appreciate that the floating rate system works very well.
Posted by: Jack Miller at December 13, 2005 08:43 AM
Certain prices rise faster than general inflation. It's pretty clear product prices fall, and I can name other things that have fallen eg air travel has fallen massively (it costs no more, in UK £ to fly to America now than it did in the late 60s, at which point the pound was worth 4X what it is now).
Product prices fall and particularly if you take into account 'hedonic adjustments' ie adjustments for quality. The 1981 IBM PC, which cost $3k, has less power than the chip in my phone now-- and of course I couldn't buy a mobile phone in 1981.
Over to rising prices. Health care and higher education are classic example of goods whose prices rise faster than inflation because they don't get the same productivity increases that manufacturing does. One of the reasons the government's share of total national income tends to rise over time.
Of course I would argue that the health care we get now is much better than it was 30 years ago-- you have a far higher chance of surviving many diseases, hip transplants and heart bypasses are commonplace, etc. Similarly in higher education there is no comparison between the environment we had (no computers) and the kind of computer aided searching and downloading a modern student can do-- I spent half my life in the library photocopying articles from journals.
Housing is a 'positional good'. If you want a nice house, odds are so does everyone else. This is an area, too, where zoning has been used to increase prices-- it's much harder to get new developments done, in most places, than it was 30 years ago. Anti-growth amendments are everywhere.
Posted by: John at December 13, 2005 09:32 AM
The Gold Standard (or 'Gold Exchange Standard') is simply a special case of Fixed Exchange Rate Regimes in general?
So the credibility problem extends to all such regimes. Witness Argentina in 2002.
Doing away with a currency, as happened in the case of the Euro, in theory reduces those costs of credibility.
*however* in the case of the euro, the mechanisms still exist to relaunch national currencies. My own theory is that Italy, which is teetering on Depression as a result of its entry into the Euro and the political intractibility of its budget and labour problems, pensions etc., will be the first country to leave the Euro.
Italians hate the Euro-- it has been shown that restaurateurs and small merchants took the opportunity to raise prices by c. 10% at the time of the Euro conversion. And in Italy, restaurateurs and small merchants are utterly critical to the way of life, so everyone is hurting (only in Greece was there more flagrant abuse).
Posted by: John at December 13, 2005 09:36 AM
A comprehensive history can be found in 'Money, whence it came where it went' by J K Galbraith. This book gives history and evolution of currency regimes very comprehensively.
Posted by: Pankaj at December 13, 2005 09:59 AM
Keynes himself wrote several essays (included in "Essays in Persuasion") on the horrible negative arbitrary and unfair effects he anticipated if Britain returned to the Gold standard with a currency re-valuation after WW I.
One of the most interesting points he makes is that Gold is far from a stable index in terms of purchasing power. Gold took a huge purchasing power hit after WW I.
The Gold-bugs argued that Britain owed pre-war debts in nominal pounds and honour required paying those debts in equally valued currency (which devalued a lot over the war). But that put a huge burden on people who had borrowed newer money. And in general would stifle the economy. He proved to be right. Britain ignored his warnings and went on the Gold Standard. After economic disaster, decoupled from gold and un-did the re-valuation.
Posted by: Matt Rochlin at December 13, 2005 02:46 PM
John a 100% reserve gold standard is not a form of price fixing.
Matt The fact that it was wrong for britain to go back on the gold standard at 4 pounds to the ounce does not mean that they could not have done so at a more resonable price that took the intervening inflation into account.
My general reply to the essay above is here http://anglosphereunionnow.blogspot.com/2005/12/gold-standard-defense.html
Posted by: Steph at December 13, 2005 09:08 PM
Another thing that happened in the U.S. was that when the country went off the gold standard in the 1930s, they confiscated all the gold. It became illegal to own gold in the U.S.! I believe that continued until the 1960s or 1970s.
I'm not sure but I think the reason is that going off the gold standard would cause an immediate increase in the dollar price of gold. That's why they were going off, because gold had an artificially low price. But if you do this, then all of the gold owners, who are mostly rich people, will get an enormous windfall as their gold holdings shoot up in value.
This was, I think, seen as politically unacceptable. During a time of great economic hardship the U.S. government could not take a step that would amount to an enormous wealth transfer into the hands of the rich. So the government was forced to confiscate the gold and make owning it illegal, in order to be able to take the political step of suspending convertability of dollars to gold and going off the gold standard.
Does anyone who is more expert on the topic know if this is the real reason that gold confiscation occured? If so it is another argument against the gold standard, that it puts governments in this position where they are more likely to engage in confiscatory behavior.
Posted by: Hal at December 14, 2005 04:10 PM
Not possible, you say? Think again-- one of these days (probably not anytime before 2050, but certainly by 2100) we'll be mining the moon and asteroids for metals, and there's a LOT of material out there to be found.
I would be flatly astonished if this came to pass.
The energy cost of mining metals from asteroids is vastly greater than the energy cost of mining metals from magnesium nodules from the ocean floor. Yet no one is mining magnesium nodules, nor is there much prospect of such in the near future. Part of the reasons are sociopolitical (there are all kinds of treaty barriers to ocean mining) but most of the reasons are economic.
But while your technological speculation is implausible, your economic argument is solid enough. William Jennings Bryan failed of election largely because the invention of the cyanide process, together with the discovery of the South African gold fields, resulted in a large increase in gold production, and in inflation of currencies on the gold standard. While I have to chuckle at the idea that asteroid mining will produce a similar influx of gold, the possibility that some other technology will increase the gold supply is real enough.
Posted by: Kent at December 14, 2005 07:32 PM
Alaskan gold production also came on line at the end of the 1800s.
Posted by: Barkley Rosser at December 15, 2005 11:33 AM
Question: What form of payment does a gold dealer accept?
Answer: Cash or Credit Card - LOL
If you want to see how much things are "getting better" in respect to the dishonest Central Bank inflationary debauching scheme hit the URL I provided.
"A fiat money standard only works when everybody believes in the overall fiscal and monetary responsibility of the major world governments and the relative exchange rate of the fiat currency is fairly stable." Sound familiar?
With regard to the first sentence of this post "How the gold standard contributed to the Great Depression" Gold NEVER caused the depression, it was misguided economic policies that cause the depression and forced the removal of the Gold Standard.
Peter Schiff writes: "The mistake made by the Fed during the 1920's was expanding the supply of money and credit too rapidly. However, as increasing productivity prevented consumer prices from rising, the Fed was unconcerned
about the inflation it was creating. Instead, the excess money and credit that spilled into financial and real estate markets caused asset prices to rise, which resulted in claims of a 'new era' (sound familiar?). The bust of
1929 led to the Great Depression of the 1930's not as a result of Fed tightening, as Bernanke claims, but due to the misguided economic policies of the Hoover and Roosevelt administrations. "
Melchior Palyi from his book "Twilight of Gold, Myths and Realities." He writes, "The chief reason for the financial confusion in the late 1920s, as in similar eras of the past, was the credit inflation. Combined with stable
price levels, it generated a sense of security and an overestimation of the expansionary potential. This misled a dynamic society into reckless speculative ventures on an unprecedented scale. Believers in monetary
stability were carried away by their wishful thinking."
Posted by: The Nattering Naybob at December 19, 2005 05:22 PM
Generational wisdom handed down from many generations is that gold is useful during times of uncertainty. David Fisher identified long price waves over the past 800 years that end with the type of crescendo rise in prices we are again seeing, and the associated disorder in the social fabric and the economy.
And I'm sitting here with my gold increasing in value while the dollar declines. Not because the dollar has been protected by government fiscal responsibility, but because of the same inflation in money supply seen in the years leading up to 1929.
Posted by: ron at August 4, 2006 07:19 PM
Currency as a brand
I recently discovered that one of the unintended sideeffects of advertizing and publishing is that it often promotes a thirdparty brand / trademark even when there's no incentive to do so. For example bed & breakfast establishments often spend a lot of time promoting the town / city they are located in.
Then I realized that the same is true for currencies on a much larger scale : An ad that says "Coke only $1" say something possitive about the dollar, even if many people know it is not the cheapest price. Considering how many ads, publications and even conversations refer to one currency or another, I think the free media exposure they get are of the order of half a trillion dollars a year. This means the brand equity of the dollar is comparable in magnitude to the reserves backing it. It explains why a branded piece of paper can trade to such a large premium over a blank piece of paper.
Posted by: Nic Roerts at August 11, 2006 05:36 AM