Home > Causes of Inflation, Commodities, Gold, Good One, Investing, Stock Market > After All, Tomorrow Is Another Day

After All, Tomorrow Is Another Day

“I can’t think about that right now. If I do, I’ll go crazy. I’ll think about that tomorrow.” – Chancellor Merkel

Actually, that quote is from Scarlett O’Hara in Gone With The Wind, but it may as well apply to the EU leaders who announced today that the much-heralded summit meeting this weekend would be delayed until ‘no later than’ next Wednesday. The cited reason is that there are still some issues to be worked through before “final agreement” can be reached. I don’t know about you, but that sounds like the NBA labor dispute. If they knew when they were going to reach a final agreement, then they would have reached an agreement. Methinks this is more an admission that the issues aren’t so easily brushed aside as the early-week rumors of a resolution suggested. The good news is that this means we don’t have to trade down into the weekend and then up on Monday (or up into the weekend and then down on Monday) on the basis of expectations set and dashed.

Europe is still hogging the headlines. The Philly Fed index was great, printing +8.7 versus -9.4 expected (-17.5 last month), on strength in New Orders and Shipments. The Number of Employees subindex, curiously, declined. This has been a wild year for Philly Fed. In March it reached a 27-year high at 43.4; in August it reached a 3-year low at -30.7. Remember that this is a relative-strength indicator: respondents describe how conditions are relative to the prior month. The huge swings show great instability in the rate of growth, due to everything from tsunamis and nuclear meltdowns to Middle Eastern riots to large-scale government intervention. I wonder if the significance of Philly Fed right now is less about its level and more about its volatility. It must be hard to make long-term investment decisions in a business when conditions are changing so violently from one month to the next.

Philly Fed has not given much insight recently, but I wonder if the volatility implies something about visibility?

This may also be the reason behind the strong TIPS auction. While the level of real yields is awful, part of TIPS’ appeal is in the option-like character of inflation. That is, if you are wrong owning TIPS here compared to nominal bonds, then you lose a little bit (we’re not going to have -5% inflation for 30 years, and even if we did you’d get par back), but if inflation develops there’s no reason it can’t be +5% or more. So again, the uncertainty about the economy may contribute to the bid. Put another way, there may be interest from nominal bondholders who are more interested in getting a 1% real yield with certainty than some chance of a +2% real yield and some chance of a -3% real yield over the holding period of the bond.

On the flip side of inflation, copper was killed today, falling more than 5%. “Doc Copper” is 34% below its highs (see Chart) and at the lowest level in a year. I think copper doesn’t have the leading indicator character that it used to have (at least in legend), and I think that it isn’t likely to stay down with the world awash in cash (M2 is still rising at 15.2% annualized over the last 26 weeks), but it is a warning sign that should be recognized.

Doc Copper is feeling a little run-down himself these days.

It may seem odd, after looking at copper, to address the question, “Are Commodities In A Bubble?” True to form, perhaps, I am less interested in the answer than in a way to attack the question. Commodities are not like capital market assets such as stocks and bonds which represent an ongoing source of value and a stream of income. A bond pays a coupon or interest at maturity; a stock pays dividends or (hopefully) increases in value over time as the company retains earnings. Because these instruments have definable cashflows, even if they are speculative, they can be analyzed with tools such as net present value analysis or a dividend discount model. But commodities have no ‘yield to maturity.’ You get no dividend from Zinc. And that means we can analyze the current value of a commodity with respect to its supply and demand but otherwise relative value tools are scarce.

But wait, analyzing the current value of a commodity with respect to its supply and demand is already done through the mechanism of price discovery. In transparent markets, it is hard to argue that any particular price represents in itself a ‘bubble’ or ‘deep value.’ Unless there is some reason that prevents the market from clearing normally, the spot price will be driven by the utility of the good for the purposes of consumption.

(As an aside, the demand from commodity index funds that acquire risk through futures contracts does not, cannot, affect the spot price in a meaningful way. This is because for every long contract there is a short contract, and commodity index funds generally roll so as to never take delivery. This may affect futures prices and contangoes/backwardations, but not spot prices. On the other hand, ETFs that hold physical commodities may, by removing commodities from circulation, affect the spot price over time. But this is not a key part of what I am talking about today.)

So what should we mark commodities against? I choose: money.

Commodities, by their nature, should have a zero real return over (a long period of) time. A pile of copper remains a pile of copper. A bar of gold remains a bar of gold. Thus, if the price of a commodity changes over a long period of time, it is probably more accurate to say that it is the unit of account that is changing value. An ounce of gold doesn’t go from $500 to $1000; a dollar goes from 1/500th of an ounce of gold to 1/1000th ounce of gold. This is, of course, the argument of hard-money adherents. I don’t understand why only gold will do – and don’t want to conduct that argument – but the general point is reasonable.

So we want to compare commodities over time to money. Is the price of commodities out of line given the amount of money in circulation?

This is a harder question to answer than it sounds like. The absolute amount of M2 money has grown enormously over a long period of time. In 1959, M2 was $298bln. In 1989, M2 was $3.2trillion. In 2009, $8.5 trillion. Obviously, commodities prices have not grown anything like that amount. This is because the amount of money must grow over time to maintain a constant price level, if GDP is to keep rising. Remember MV≡PQ. If velocity is stable, and prices are stable, then M must grow with output. Of course, velocity is not stable, although M2 velocity happens now to be back in the general range it inhabited from 1960-1990. And of course, in principle, the Fed doesn’t want prices to be precisely stable, either, since a small amount of inflation is seen to have much lower costs and risks than a small amount of deflation. So if we plot M/Q, we expect to see it rise slowly over time. In fact we do, as the chart below shows (incidentally, don’t be concerned about the units on any of these costs. They are artificial; it’s the trend that matters. For example, the chart below shows how many dollars are in circulation for every dollar of output, but the axis units will be different if I use, say, 1982 dollars rather than 2005 dollars as we have here).

M2 compared to real GDP

So again, charts of commodities ought to look something like the chart above. There is more money in the system, but it’s really how much more money there is relatively since a larger economy needs more money just to keep prices stable.

I compare “economy-adjusted money” to commodity prices in the chart below using the CRB (which I use mainly because it has a long price history). I think a case can be made that commodities may have been ‘bubbly’ in the 1970s and perhaps in 2008 they were starting to get there, but it’s hard to look at this chart and see a big bubble now.

CRB compared to economy-adjusted money

Now, certain commodities might be. This is perhaps not as good a way to compare value for individual commodities because for them we also need to consider more microeconomic demand and supply and, especially, long-term changes in the supply dynamic, but we can get an idea of general trend. Live Cattle and Wheat certainly don’t look bubbly. Note that if we didn’t adjust Live Cattle, it looks very much like the M2/GDP chart: it slopes upward and to the right. Ditto Wheat, although to a lesser extent. Adjusted by “economy-adjusted money,” both of these look about fair.

Wheat and Live Cattle (chosen for random illustration)

Crude oil between 1986 and 2003 was pretty stable, and started to look expensive thereafter. To be fair, there are some questions about long-term supply in oil (Peak Oil and all that). I won’t say anything about Gold, because whatever I say makes somebody mad. You judge. I’ve written before about how it makes some sense to think about Gold as an in-the-money straddle on inflation.

Gold and Crude Oil have more interesting stories (and they always do).

Finally, and perhaps surprisingly, I threw in two financial assets that are often thought  of as inflation hedges for different reasons. The chart below shows the adjusted ratios of stock prices and median existing home sales prices to economy-adjusted money. Home prices look pretty reasonable on a long time scale; stocks less so, but of all of these charts I have the least intuition about how stocks should look through this lens.

Is this a reasonable way to 'deflate' financial assets?

So my overall perspective is that the symptom of high prices of certain commodities and of commodities in general in no way should produce a diagnosis of a bubble in commodities. Feel free to disagree and comment…I’m not saying this is the final word on the subject!

  1. October 20, 2011 at 11:49 pm

    Why wouldn’t hedgers of trades with commodity funds on futures prices use spot commodities if the relationship between the futures price and spot prices moved out of line with arbitrage pricing? And if that were the case, couldn’t futures prices then force spot prices higher as that arbitrage opportunity was eliminated? Cf: Hull & White, like Chapter 1 or 2, no?

    • October 21, 2011 at 12:22 am

      Let’s abstract from the fact that most commodity funds can’t actually take delivery of commodities. Let’s say XYZ fund takes delivery of wheat, being long in the physical, and thereby tracks the spot price of wheat while pushing demand for the physical higher. What happens next? Whether it’s wheat or oil or something else, it has no value except if it’s consumed or transformed. Eventually, the fund will have to sell it back into the physical market because it wasn’t actually the end user. So while you could have a temporal dislocation, higher prices today will imply lower prices tomorrow. And vice-versa, which is why so many guys took delivery of oil and sat it in tankers around the world. But it didn’t change the spot price permanently!

  2. October 21, 2011 at 1:37 am

    Several comments:

    1. What do you mean when you say for every long contract there is a short contract. Are you saying that one person is long therefore the counter party is short?

    2. Leaving aside for now whether stockpiles show this is happening, spot prices can clearly be effected by steep contango via arbitrage that allows a profitable buy and store of the physical commodity. Spot prices would rise to a zero arbitrage level.

    (I’ve just read Michael Thorson’s comment and it is probably much along the same lines.) So let’s extend this. Why should we assume that the arbitrager wins the bid? So imagine an arbitrage opportunity exists. You would expect arbitragers to bid up to the zero arbitrage price. But let’s also suppose that people *requiring* physical delivery are bidding in the market. They must match that bid (at least). Thus you would get spot prices rising without the arbitrager/speculator actually taking physical delivery.

    3. “Unless there is some reason that prevents the market from clearing normally, the spot price will be driven by the utility of the good for the purposes of consumption.”

    I don’t follow softs but for metals one thing you may or may not be aware of is artificial supply constraints imposed by metals warehouses. This has been written about for 12 months or so (FT Alphaville being a good source of regular articles about it). So we see rising prices AND rising stockpiles (contrary to econ 101 supply/demand analysis) but the stockpiles are not necessarily accessible (in a timely manner) to those requiring physical delivery.

    4. On commodities generally this is 7 months old but may be of interest:


    5. re: M2, an alternative view


    • October 21, 2011 at 6:11 am

      Why do you guys keep asking about the part of my comment that’s already been argued to death? 🙂 To answer Wildebeest (thanks for posting such a thoughtful note!):
      1. yes
      2. similar answer to Mike’s…this can affect today’s price, but not price in the medium-term (the important exception being where someone actually does take the commodity off the market semi-permanently, as when an ETF or a sovereign wealth fund takes and stores a metal). Eventually that commodity needs to find its way to someone who actually needs it – that’s whence it derives its value.
      3. I’m aware of that…I designed a metals arbitrage program for one of the banks I worked for. You clearly have a great understanding of this market – and I think where we disagree (and I am not even sure I can call it that) seems to be on the short-term versus long-term effect on spot prices. Unless there is an increase (or decrease) in permanent aggregate storage, this doesn’t change the long-run supply. It just changes when the supply is actually delivered to the end user.
      4. and 5. Thanks for the links! I will read them with interest. Still trying to figure all this out.

  3. Jim H.
    October 21, 2011 at 8:52 am

    ‘I won’t say anything about Gold, because whatever I say makes somebody mad.’

    Okay, I’ll say it. In terms of your ‘economy-adjusted money,’ gold looks like it’s double-topping.

    Commodity valuation is a problem I’ve pondered for years … and it’s frustratingly elusive. The CRB and bonds used to be exhibit a fairly reliable inverse correlation. But when safe haven and deflation fears rule, bond correlations get turned upside down.

  4. October 21, 2011 at 3:48 pm

    1. Ok that’s what I thought you must have meant. That’s fine.

    2. In my example the commodity gets taken off the market because the industrial user (entity requiring physical delivery) wins the bid. So where I think we seem to differ is that we both agree that an arbitrage situation can raise the spot price but that you believe this will always lead to further stock coming back on to the market at some future date — therefore pushing the spot price at that time back down. I say that’s not necessarily the case because the existence of an arbitrage puts a price floor but doesn’t mean the arbitrager wins the bids — the commodity gets delivered to an industrial user and doesn’t get added back to stock for sale at a later date.

    3. Upon further thought this could tie in with #2. For example (and I’m restricting my analysis to metals) you could buy on spot now but you may not get your stuff out of an LME warehouse for 3-6 months. So even if the arbitrager were to win the bid they could rely on the fact that the stuff is going to be warehoused for 3-6 months (to take advantage of arbitrage with e.g. the 3 month contract). In other words the storage part of the equation is kind of naturally built in.

    Incidentally buyers will often pay a premium to spot to buy directly from a refinery in order to take spot delivery (due to LME warehouse delays).

    4. Here’s another link discussing commodities in a historical context that you may be interested in reading at some stage (10 pages). Personally I find the bank of japan analysis (linked yesterday) more robust.


  1. October 22, 2011 at 9:05 am

Leave a Reply

%d bloggers like this: