About those high gold prices

As far as I know there is really only one respectable argument that inflation expectations are approaching dangerous levels in the US.  We know that 5 year TIPS spreads are low, and we know that the near to medium term consensus inflation forecast is low.  We know actual inflation is low and falling.  But then there are those gold prices.

I’ve never been convinced that the high gold prices were signaling US inflation fears.  One problem is that gold prices are set in a global market, so it’s not clear why we should assume they are forecasting high inflation in the US, rather than the eurozone, Japan, India or China.  India has generally been the world’s largest buyer of gold.  Furthermore, most other metals prices have also been soaring, presumably due to rising demand in the developing world.

First let’s look at the supply side of the equation.  We can’t directly observe supply curves, only equilibrium points.  But nonetheless the recent price and quantity data suggests that supply may be falling.  We know that gold prices have soared in terms of all major currencies, and yet gold output continues to decline rapidly:

Note that production peaked at 2600 tons, and fell to 2260 tons in the latest year available (2008.)  I’ve read articles suggesting that most of the world has now been thoroughly explored, and it is felt that most of the major goldfields have already been discovered.  Producers are having to concentrate on less high quality ores.

Set against that bleak supply picture, we have the following news from the FT about gold demand in China:

Gold imports into China have soared this year, turning the country, already the largest bullion miner, into a major overseas buyer for the first time in recent memory.

The surge, which comes as Chinese investors look for insurance against rising inflation and currency appreciation, puts Beijing on track to overtake India as the world’s largest consumer of gold and a significant force in global gold prices.

The size of the imports – more than 209 tonnes of gold during the first 10 months of the year, a fivefold increase from an estimate of 45 tonnes last year – was revealed on Thursday. In the past, China has kept the volume secret.

“Investment is really driving demand for gold,” said Cai Minggang, at the Beijing Precious Metals Exchange. “People don’t have any better investment options. Look at the stock market, or the property market – you could make huge losses there.”

Beijing has encouraged retail consumption, with an announcement in August of measures to promote and regulate the local gold market, including expanding the number of banks allowed to import bullion.

Shen Xiangrong, chairman of the Shanghai Gold Exchange, who disclosed the import numbers, said uncertainties about the Chinese and global economies, and inflationary expectations, had “made gold, as a hedging tool, very popular”.

The rise in Chinese demand could further inflate gold prices. Bullion hit a nominal all-time high of $1,424.10 a troy ounce last month. But adjusted for inflation, prices are far from the 1980 peak of $2,300.

“The trend is undeniable – gold demand in China is rising rapidly,” said Walter de Wet, of Standard Bank in London.

.   .   .

Chinese total gold demand rose last year to nearly 450 tonnes, up from about 200 tonnes a decade ago, according to the World Gold Council, the lobby group of the mining industry. Analysts anticipate a further leap this year, putting the country within striking distance of India’s total gold demand of 612 tonnes in 2009.

Those Chinese consumption estimates are rather large when set against a world production total of 2260, and falling.  I seem to recall someone pointing out that Asian gold demand couldn’t be the problem, because the totals were fairly stable.  But that confuses shifts in demand with movements along a demand curve.  When world output is falling, it is necessarily true that total quantity demand will also fall.  If you confuse demand and quantity demanded, it will never look like higher demand is pushing up prices when output is declining.  For instance, consider the following scenario:

1.  The world supply of gold is shifting to the left, and is expected to continue doing so.

2.  Chinese demand is rapidly shifting to the right, and is expected to continue shifting right as Chinese wealth rises dramatically.

3.  Indian demand is shifting right, but much more slowly.

Here’s what I would expect.  Global gold prices would be expected to rise strongly over the next few decades.  If markets are efficient and nominal interest rates are very low, the expected future rise should also raise current gold prices.  The intertemporalarbitrage necessary to make this happen would be provided by gold speculators who buy up lots of gold, driving current prices much higher.  Because Indian demand is increasing only modestly, yet remains the world’s largest, the higher price causes India to slide up and to the left along their demand curve, to a higher price and a lower quantity demanded.  People might wrongly assume that Indian demand had declined, whereas in fact Indian demand was still increasing (due to income growth) but the price was increasing so fast that quantity demanded declined.

The fact that Chinese quantity demanded increased rapidly despite a huge rise in gold prices suggests that the Chinese demand curve is moving to the right at a truly explosive rate.  Indeed in this model it would be possible for future expected Chinese demand to explain 100% of the world’s recent gold price rise, even if the current quantity demanded were falling slightly.  Speculators might be holding gold stocks in readiness for future expected Chinese demand.  But of course Chinese consumption (Qd) is rising fast despite the high prices.

The real price of gold is still much lower than in 1980, a time when China was not a factor in the world gold market, and also a time when lots of new gold fields were yet to be discovered.  I can easily understand how savvy hedge fund managers could see the China boom as being very bullish for gold:

The surge in gold imports to China bodes well for some of the world’s biggest hedge fund managers, including David Einhorn of Greenlight Capital and John Paulson of Paulson & Co, who have invested heavily in bullion, and top miners Barrick Gold of Canada, US-based Newmont Mining and AngloGold Ashanti of South Africa.  [Also from the FT article.]

Again, I don’t doubt that there are individual investors fooled by news stories of massive monetary base increases and huge deficits, who think that high inflation is just around the corner.  Not many average investors know that the same thing happened 15 years earlier in Japan, with no inflationary consequences.  But I believe the best point estimate of the market’s consensus inflation forecast comes from the CPI futures market, as well as the TIPS spreads in the T-bond market.

One interesting question is why Chinese inflation is rising sharply.  It’s clearly not due to any rise in US inflation; rather the most likely explanation is the Balassa-Samuelson effect.  I’ve been predicting a huge appreciation in the real value of the yuan for quite some time, and so far I’ve been right.  A recent article in The Economist pointed out the real yuan is up about 50% since 2005, with about half the increase being nominal appreciation and the rest representing higher inflation in China.

Update:  Commenter David Pearson sent me the following information:

Demand and supply statistics

Gold Demand and Supply – Q2 2010


The World Gold Council recently published the latest issue of Gold Demand Trends for Q2 2010, which suggests demand for gold for the rest of 2010 will be underpinned by the following market forces:

  • India and China will continue to provide the main thrust of overall growth in demand for the remainder of 2010, particularly for gold jewellery.
  • Retail investment will continue to be a substantial source of gold demand in Europe.
  • Over the longer-term, demand for gold in China is expected to grow considerably. A report recently published by The People’s Bank of China and five other organisations to foster the development of the domestic gold market will add impetus to the growth in gold ownership among Chinese consumers.
  • Electronics demand is likely to return to higher historic levels after the sector exhibited further signs of recovery, especially in the US and Japan.

Note that none of the major factors driving future gold demand is associated with inflation fears in the US.  The report also indicated that investment demand has risen strongly in 2010, but this is consistent with my argument that higher future Chinese demand, coupled with slow growing supply, would lead speculators to buy gold in anticipation of future price increases.

Here is another way of thinking about the problem.  Suppose that the Chinese yuan is expected to double in real terms over the next 25 years (which I think is quite likely.)  Also suppose that gold prices in dollars are expected to rise at the risk free interest rate (on T-bonds.) In that case gold prices in yuan terms might be expected to show little change over the next 25 years—which would lead to a much higher quantity demand as China becomes much richer.  That future expected Chinese demand (partly driven by the Balassa-Samuelson effect combined with Chinese exchange controls) could very easily be underpinning current investment demand in the US.

PS.  The article above links to “Gold Demand Trends,” which shows a wealth of detailed information about the gold market.  It seems gold ouput has risen a bit since the 2008 figures I got from Wikipedia.  But that doesn’t mean the supply curve is increasing.



28 Responses to “About those high gold prices”

  1. Gravatar of David Pearson David Pearson
    4. December 2010 at 09:07


    Not sure of your source. The World Gold Council produces supply and demand stats. They show supply — net of hedging — up around 15% in 2009 and another 8% in YTD 2010. Supply is also up excluding hedge reversals.

    Keep in mind that supply net of hedging is the key metric as far as the impact on price is concerned. Miners began buying back forward sales in 2005, and it has continued until this year.

    The boom in Chinese imports is concentrated in investment demand if the WGC figures on Jewelry are correct. Why would one hoard gold instead of currency as a store of wealth? There is only one reason, and that is to hedge against against a decline in the purchasing power of the currency. With Chinese inflation accelerating and real deposit rates negative, it is no wonder why the Chinese are choosing to hedge.

    While quantity demanded is not demand, the biggest change in quantity demanded since the gold price started spiking — from sales to purchases — is by Central Banks. Clearly, they are purchasing more at higher prices, and this is an indication of a curve shift. There is only one reason for central banks to favor gold over dollars as a store of reserves, and that is uncertainty over the purchasing power of the dollar.

    Yes, the gold price has been rising in other currencies as well. Perhaps that has something to do with the likely possibility that those countries (Japan and the Eurozone) might choose monetize high fiscal deficits on a long-term basis. In the case of the dollar-bloc countries (including China), the gold rise in local currency terms indicates a low probability of the country abandoning the dollar bloc with a maxi-reval.

  2. Gravatar of David Pearson David Pearson
    4. December 2010 at 09:20

    Attached is the WGC link. Click on “Gold Demand Trends”, mid way down the page to see the latest (Q3) report.


  3. Gravatar of Luis H Arroyo Luis H Arroyo
    4. December 2010 at 09:40

    Balasa-Samuelson, catching up. Two sectors, productive-improductive. Yes, quite possible indeed. But, I can´t imagine why the chinese authority have any interest in openning the gold market for everybody.

  4. Gravatar of rob rob
    4. December 2010 at 10:44

    It’s not gold prices that worry me; it’s oil prices. You’d think the world economy was kicking ass if you just looked at oil.

  5. Gravatar of Benjamin Cole Benjamin Cole
    4. December 2010 at 11:06

    Superb essay by Scott Sumner–precisely my conclusion after reviewing world gold markets a few months back.

    If Chindia demand for gold continues, we may yet see record gold prices–and it has little to do with US monetary policy.

    The US is a smaller player in world gold markets.

  6. Gravatar of TravisA TravisA
    4. December 2010 at 11:11

    Scott, I don’t get this line:

    “Also suppose that gold prices in dollars are expected to rise at the risk free interest rate (on T-bonds.) ”

    What’s the basis of that assumption? How is this consistent with the earlier part of the post that is explaining a big *rise* in gold prices?

  7. Gravatar of ssumner ssumner
    4. December 2010 at 11:16

    David, Thanks, I added an update that reflected that information. A few comments:

    1. The link you provide mentions 4 factors driving future gold demand higher. None are related to inflation fears in the US.

    2. I agree that there is fear of inflation in China, but that is exactly my point, it is fear of Chinese inflation, not US inflation.

    3. I agree that central banks are buying gold. The article you provide indicates that the motivation is to diversify their portfolios. Note that I am focusing on the question of whether the market expects higher inflation. The motivations of the President of the PBOC tells us nothing about market inflation expectations. His decision is a political decision, indeed he knows no more about the future path of American inflation than I do. I’d guess the dollar will fall against the yuan over time even if we have low inflation. So it’s a good idea for the Chinese central bank to supplement their dollar holdings with gold. I’d do the same.

    4. The data supplement also mentions large Middle Eastern jewelry demand, another factor unrelated to US inflation.

    5. The population of Asia is really, really big, and getting richer at a fast rate. They also have a gold hoarding tradition. Even the Bangladesh central bank just bought 10 tons of gold. If we really are nearing a peak gold scenario, then the gold market may stay tight for quite some time, even with low inflation in America.

    Luis, That surprised me too, and Mao must be rolling over in his grave.

    rob, The world economy IS kicking ass–remember that most people are Asians. It’s increasingly becoming true that the world economy is the Asian economy. And Asia has been growing fast (all the growth rates for 2010 are in the 5% to 10% range.)

  8. Gravatar of Muckdog Muckdog
    4. December 2010 at 11:21

    I don’t think we have seen the blow off top in the gold bubble yet. Like we saw in stocks and real estate, the last move in the bubble is parabolic.

    Just don’t be the last one out.

  9. Gravatar of scott sumner scott sumner
    4. December 2010 at 11:39

    Benjamin, Thanks.

    Travis, It is completely consistent. Big asset price rises are almost always unexpected. Expected increases are limited by the efficient market hypothesis, which says that the expected gain in one asset prices will be (adjusted for risk) about the same as another investment asset.) Obviously gold is risky, but it has a negative correlation with many other risky assets, which makes it an attractive investment. I believe the term they use in finance is that it has a negative ‘beta.’

    So if new information about Asian gold demand causes next year’s expected price to suddenly soar to $2000, and if interest rates were 5%, then the current price should immediately rise to $1900, and then would be expected to gradually rise from that point forward.

    Muckdog, Thanks for the valuable investment tip—how much do I owe you?

  10. Gravatar of David Pearson David Pearson
    4. December 2010 at 12:14


    In its supply/demand publications, the WGC talks about “investment demand” for gold. Elsewhere on its site, it clearly argues that “investment demand” should be driven by inflation hedging.


    From the statistics, it is clear that “institutional” and central bank demand for gold has been the swing factor in demand in the past few years. Again, quantity demanded is not demand, but increased demand at higher prices is clearly a sign of a demand shift. In contrast, Indian jewelry and industrial gold quantity demanded has fallen as prices have risen. Perhaps the price rise masked a rightward shift in demand, and perhaps not.

    With regard to “portfolio diversification” intentions by central banks, this is a vague construct. Two things have happened:

    1) the European central banks sold gold in the late 90’s and the early 2000’s. Quite a bit of it. They did so clearly because they thought gold was an unnecessary “relic” in a time when the dollar could be counted on as a reserve currency. This coincided with the Fed’s “opportunistic disinflation” strategy. Sometime after the shift to asymmetric policy in 2002, the European CB’s decided to slow down sales and then stop altogether.

    2) We cannot know the PBOC’s intentions. Certainly they would not publicly trash the dollar as it would hurt the vast majority of their holdings. We do know that the CB’s of emerging market reserve accumulating countries began to buy gold after the financial crisis. This may have been a sudden recognition of portfolio theory, or it may have been a reaction to QE. The evidence from some recent public central bank/FinMin comments (Russia, Argentina, Brazil) favors the latter explanation, not the former.

    On the institutional side, the “savvy” hedge fund managers are on record as buying gold to hedge against future inflation. These include: John Paulson, George Soros, Paolo Pellegrini, Eric Mindlich (of Eton Park), Kyle Bass, and, yes, as you pointed out, David Einhorn. Below is Einhorn’s succinct investment thesis on gold:

    “Our guess is that if the chairman of the Fed is determined to debase the currency, he will succeed. Our instinct is that gold will do well either way: deflation will lead to further steps to debase the currency, while inflation speaks for itself.”


    In two waves, after the adoption of “extended period” policy in 2003 and QE in 2008, European central banks stop selling, and institutions and reserve accumulator central banks start buying. These were big shifts, and they led to a dramatic price rise. Why did this occur? The simple explanation is the best.

  11. Gravatar of OGT OGT
    4. December 2010 at 12:27

    This would make a good working paper topic.

    The Economist version of “real” exchange rate doesn’t make much sense to me. I was under the impression that one should include productivity in calculating a real exchange rate. For example, if nominal appreciation plus inflation differentials over a given period was 50%, but the net relative productivity growth was 48%, the actual “real appreciation,” as expressed by something like unit labor costs, would be only 2%. Is that incorrect?

  12. Gravatar of TravisA TravisA
    4. December 2010 at 12:55

    Thanks for the explanation, Scott. Extremely cogent, as usual.

  13. Gravatar of Mattias Mattias
    4. December 2010 at 14:08

    It seems everyone agrees that gold should go higher. That’s usually not a good sign. I wouldn’t be surprised if we are much closer to the top than most people think. I think the strong investor demand is a real warning sign. When the stock market gets closer and closer to its old highs and there’s still no inflation, lots of investors will get rid of their gold.

  14. Gravatar of JP Koning JP Koning
    4. December 2010 at 19:11

    “As far as I know there is really only one respectable argument that inflation expectations are approaching dangerous levels in the US. We know that 5 year TIPS spreads are low, and we know that the near to medium term consensus inflation forecast is low.”

    Anyone looking for evidence of currency debasement in TIPS spreads hasn’t figured the game out yet. Nowadays, those expecting inflation act out their fears in asset markets, not consumer product markets. Look at the rise in the S&P and commodity futures markets, and the plunge in fixed income/municipal bonds for a hint. TIPS are a red herring.

  15. Gravatar of Mark A. Sadowski Mark A. Sadowski
    4. December 2010 at 19:39

    If I had read this in 2008 I wouldn’t have sold my fathers gold coin collection. 😉

  16. Gravatar of scott sumner scott sumner
    5. December 2010 at 06:08

    David, I see the central bank pattern differently from you. The European central banks are not big buyers. Yes, they have stopped selling, but that may be because they want to hold some minimum gold amount.

    Much of the gold demand is coming from central banks in growing Asian countries, which is entirely sensible, as their currencies will appreciate against the dollar regardless of whether the US has inflation or not. The fact that they also hold a huge quantity of Treasury bonds (much more than gold) suggests they aren’t particularly concerned about US inflation. T-bonds would be a terrible investment if you feared inflation, indeed almost the worst.

    But again, I don’t think central bank purchases and sales tell us ANYTHING about market inflation expectations. They are government agencies, not the market. If I trusted the government, I favor having the Fed determine monetary policy. Since I don’t I favor having the market determine monetary policy.

    You quote a hedge fund guy as saying gold will do well even if we have deflation. But if we might have deflation, then the Fed absolutely should be doing a more expansionary policy right now, to prevent that from happening. So it would make no sense for a QE opponent to say, “see, hedge funds are buying gold, don’t ease monetary policy” if they are buying gold partly because they fear deflation.

    OGT, That is incorrect, you do not include productivity in real exchange rate calculations, just relative price levels and nominal exchange rates. The real rate is the nominal rate times the domestic price level divided by the foreign price level.

    Thanks Travis.

    Mattias, I hope you are right, I’m in stocks.

    JP Koning, That makes no sense. Inflation is DEFINED as a rise in consumer prices.

    Mark, I was thinking of buying some gold coins a few years ago when a $20 dollar gold piece (roughly an ounce) cost $500. I wish I had pulled the trigger.

  17. Gravatar of JP Koning JP Koning
    5. December 2010 at 08:33

    No, inflation is a broad decline in the value of the currency that occurs when people aggressively sell currency for X,Y, and Z, pushing down said currency’s market value. If X, Y, and Z happen to be financial assets, then the decline in the currency is not registered in the formal CPI, but in equity markets, gold markets, housing markets, art and other collectibles. Like I said, if you’re not watching these markets for indications of currency debasement and expected debasement, ie. inflation, then you haven’t figured out the game.

  18. Gravatar of StatsGuy StatsGuy
    5. December 2010 at 10:03

    Note that it’s hard to judge future gold production costs based on current mine output. I believe, right now, the marginal mines operating have a production cost of roughtly 600 dollars an ounce. At that cost, which no one was prospecting for 10 years ago (when gold was 300 dollars or less), the potential supply opens up significantly – particularly if new technologies come online. It can take 5 years or more to bring a new resource online. Observing the price increase, there is probably a longer term ceiling on gold costs, so the notion that it’s physical supply that’s the problem probably doesn’t account for most of the price changes. Gold only broke 600 dollars in price ~5 years back, and at that time prospecting was probably focusing on resources that cost 400 dollars or less to ensure a profit.

    There is another explanation for the gap between gold and TIPS – one that the radical austrians firmly believe. That is, real inflation is much higher than government reported inflation, and since TIPS are linked to govt. reported inflation, they’re not really keeping up. Thus, the true measure of inflation is gold. Likewise, one might ask WHY China is increasing investment demand? Could it have anything to do with perceived insecurity of holding reserves in US debt?

    I don’t really have an axe to grind here, but the argument you present does not completely answer the Austrian challenge.

  19. Gravatar of ssumner ssumner
    5. December 2010 at 17:20

    JP Koning, No, inflation is the rise in consumer prices. It is not the fall in the value of the dollar. Otherwise whenever the dollar rises we’d have deflation.

    Statsguy, I did address the China argument. The Chinese government is smart to invest in gold, because the value of the yuan will rise due to the Balassa-Samuelson effect. But that doesn’t tell us anything about US inflation expectations.

    Regarding whether there is inflation that is being hidden by the government, everything is relative. Whatever flaws exist in the CPI were there 10 years ago as well. In a relative sense inflation is clearly low, lower than in recent history. The people pointing to fast MB growth surely have in mind more than 1%, or even 2% or 3% inflation. Beyond that level conspiracy theories of inflation become completely implausible.

    Those are good points about gold supply. But the flow supply is extremely low relative to the stock. So it can take a awful long time for gold prices to fall to cost of production. In any case, I have an open mind on the issue of where gold prices are going–indeed I’m somewhat of a bear. I just don’t see any evidence that expectations of inflation in the US are the major factor in gold prices. Other metals prices are also soaring.

  20. Gravatar of Mike Mike
    5. December 2010 at 17:54

    The overinvestment in stocks that peaked in the dot com bubble could also play some role in the recent decreases in the supply of gold. Zeal LLC has been posting essays since 2000 at


    predicting progressively larger spikes in gold prices that should lead to a bubble sometime in the next 8 years (when EVERYBODY starts buying gold) based partially on underinvestment in mines as money flowed into stocks, which they still see as overvalued based on trailing 10 year EPS, despite currently being in a cyclical bull market. The flow of money into stocks supposedly diverted money from the large capital expenditures required to start a gold mine, and has not been corrected yet due to the long lead times required to do so.

    I would like to know what you think of their theories because although they mirror you in crediting the rise of Asia and seeing central banks as poor investors, they seem to be “fooled by news stories of massive monetary base increases and huge deficits” into thinking “that high inflation is just around the corner.”

    Maybe it’s just that they haven’t seen your site yet, and been swayed by it, as I have.

  21. Gravatar of Shailesh Dhuri Shailesh Dhuri
    5. December 2010 at 19:03

    The real game changer for Gold is the change of perception in the minds of Asian Investors, including Asian central banks, with respect to the willingness, and to a certain extent, ability of Western central banks to protect real value of their respective currencies.

    While the Western investors continue to judge the relative safety of their emerging market investments on, inter alia, the level of foreign exchange assets held by a country; for an Asian Central Bank, any increase in foreign exchange reserve, must be creating a feeling of panic. There is no “ab initio safe” currency, which one can say with fair degree of confidence is not hell bent on quantitative easing. QE will debase western currencies. Asian policy makers appreciate that, and that is the main force for whatever is going to happen to the Gold.

  22. Gravatar of TheMoneyIllusion » Do the QE opponents have ANY good arguments? TheMoneyIllusion » Do the QE opponents have ANY good arguments?
    6. December 2010 at 06:35

    […] no, for reasons discussed in this earlier post.  Every direct indicator we have of inflation expectations shows very low inflation in the years […]

  23. Gravatar of Marty Riske Marty Riske
    6. December 2010 at 08:12

    Gold represents insurance against an unlimited supply of paper currency and electronic digits.

    All currencies are declining in a trade war to reduce the export costs in other countries.

  24. Gravatar of spencer spencer
    6. December 2010 at 10:07

    Gold supply and demand is more complex.

    For example, South Africa is the worlds largest gold producer.

    But the price elasticity of supply for South African gold is negative. That’s right, as the price of gold rises South Africa reduces its production.

    That is because South Africa is not trying to maximize short run profits. Rather, they are trying to mazimize the life of their mines and the total amount of gold they will mine over the long run. Consequently they mine the lowest quality gold veins that
    will allow them to still pay their dividend and undertake the necessay investment to maintain the mines. So when the price of gold rises the supply of South African gold goes down.

    In the 1970s I was a gold analyst and did a major study of the long term supply trends for gold and non-monetary demand for industry, jewelery and hoarding in third world countries like China and India. The study concluded that over the long run the long run non-monetary demand for gold would tend to grow faster than the supply of gold.

    Thus, to keep supply and demand in balance, given the relevant demand and supply elasticities, the real price of gold would have to rise at something like a 3% to 5% rate.

    Since then several things have impacted the market — Russian sales of virtually all of its large gold stocks, other central banks selling gold, weak Japanese demand because of economic stagnation, the emergence of China as a major market and the legalization of gold imports in India and more rapid growth in Indian demand. On the supply side there have been some significant technological improvements in gold refining methods.

    Secondly, 1969 was a good base year for prices because their were no significant central bank buying or selling in that year. So the $39 price in 1969 seemed to represent an equilibrium price.

    If gold prices had risen at a 3% annual real rate since 1969 they would be almost $1,000 now and if the real rate of appreciation had been 5% the current price would be about $2,250. So current prices do not seem to be out of line with the long run fundamentals even if speculation or inflation hedging is playing a role in the recent run up in gold prices.

    P.S. the World Gold Council work brought up in earlier comments is a good source of gold analysis, and they credit growing Chinese and Indian demand for much of the recent surge in gold prices.

    PPS. the gold system as it existed in the late 19th century was not a free market system. It completely depended on central banks offering to buy all gold tendered to it at a price that was well above any foreseeable market clearing price. How else could the central banks expand their gold stocks. I give a great deal of credence to the Mundell argument that a major cause of the depression was an inadequate and poorly distributed supply of gold. Moreover, part of the problem in the 1960s was that the $35 dollar price was too low. But if we were now to return to a gold system like many conservatives seem to want, the price probably would have to be around $5,000 for it to work.

  25. Gravatar of David Stinson David Stinson
    6. December 2010 at 10:39

    Hi Scott.

    If I could summarize your argument, it is that gold price increases over the last decade and the last 18 months are not driven by an increase in monetary or investment demand but by declining annual output and growing industrial demand.

    I would offer the following thoughts.

    First, as I understand it, the key supply metric in the gold market is not annual output but accumulated supply. Most gold is not consumed, but rather held, in the form of bullion or jewellry. The supply available for trading is much higher than current output.

    Second, quite apart from the absence or presence of inflationary expectations, the price of gold is inversely related to the level of real interest rates. Real interest rates are low. Plus, there is no counter-party with gold, unlike in the case of many notionally “guaranteed” financial assets. That has value particularly in a deflationary environment.

    Third, if your argument was correct, at least in respect of the last 18-24 months, one would expect the stock prices of gold miners to have followed gold prices up. However, since September/October 2008, they have largely gone nowhere (perhaps until just recently?). They fell off a cliff in the fall of 2008 and have not really recovered, or at least not to the extent that I gather would have been expected given the increase in gold prices.

    Fourth, inflationary expectations are not, I would suggest, fully captured these days by a point estimate or a single expected value. In calmer times, not only would inflationary expectations across individuals exhibit more similarity but each individual’s estimated probability distribution of future inflation outcomes would likely be much more concentrated around the mean. Nowadays, while most investors probably think that, on balance, inflation will remain very low, I suspect that many of those same investors also believe that there is a non-negligible probability that inflation will be substantially higher. As the flailing continues at the policy level and ever larger stimulus is considered, it is also reasonable to expect that even investors who believe stimulus is justified might also recognize that such policies carry risk.

    In such circumstances, gold may trade in part like an option. The greater the range of potential likely outcomes, the greater the volatility and the greater the value. The network effect that helps to sustain the choice of a reserve currency in stable times could in unstable times help to speed a decline in value. This increases the potential future volatility in an environment in which confidence in the currency or policymakers has been undermined.

    Thus, TIPS and gold may not be sending inconsistent signals, they may just be conveying information about different elements of inflationary expectations. Put another way, you may not expect your house to burn down (the point estimate) but you will probably still buy fire insurance (the option).

  26. Gravatar of scott sumner scott sumner
    6. December 2010 at 19:15

    Mike, I am skeptical of anyone who claims to be able to predict gold prices.

    Shailesh, I hope Asian central banks aren’t only now figuring out their currencies will appreciate sharply against the dollar–the Balassa Samuelson effect has been known for decades. I’ve been saying the yuan is eventually going to 3/1 ever since the early 2000s, when it was 8.2.

    Marty, I don’t see any currency war–if we had one we wouldn’t have 1% inflation. Remember the 1970s?

    Spencer, That’s not my area of expertise, but interesting analysis.


    I agree about real rates, and have made the point elsewhere.

    I do not have any theory of gold mining stock prices, so I disagree with you there.

    Gold may be an option, but it still tells us nothing about inflation expectations. TIPS give us something concrete to work with. I think TIPS have been and will continue to be much more accurate.

  27. Gravatar of JP Koning JP Koning
    6. December 2010 at 19:19

    “No, inflation is the rise in consumer prices. It is not the fall in the value of the dollar.”

    What can I say, this comment makes me oddly happy. As long as you actually believe that, and a lot of people follow you, that means easy trading profits for me! Why am I even arguing? Tra la la.

  28. Gravatar of Velk on QE2 « Keat Yang's Economics Observations Velk on QE2 « Keat Yang's Economics Observations
    5. January 2013 at 17:24

    […] linked from Greg Mankiw’s blog). In this I share the view of  economists who disagreed. Scott Sumner for instance, argued that the increasing price of gold is not due to the fear of inflation […]

Leave a Reply