Menzie Chin on futures as predictors of commodity prices

Here is Menzie Chin (my thoughts below):
As commodity prices start rising again -- at least some -- the question of whether futures are useful indicators seems relevant. Figure 1 shows the IMF commodity price indices, as reported in the October World Economic Outlook:

Figure 1: Commodity price indices for energy (blue), food (red), agricultural raw materials (green), metals (black) and beverages (teal). NBER defined recession shaded gray, assuming recession ends in 2009M06. Source: IMF, World Economic Outlook (October 2009), data for Chart 1.16.

In a previous set of papers, Oli Coibion, Michael LeBlanc and I examined the predictive power of energy futures post and paper.
In a new paper, Oli Coibion and I update our results regarding energy futures, and metal and agricultural commodities as well, through the end of August 2008, just before the financial crisis broke out in full force. From the paper:
This paper examines the relationship between spot and futures prices for commodities, including those for energy (crude oil, gasoline, heating oil markets and natural gas), precious and base metals (gold, silver, aluminum, copper, lead, nickel and tin), and agricultural commodities (corn, soybean and wheat). In particular, we examine whether futures prices are (1) an unbiased and/or (2) accurate predictor of subsequent spot prices. We find that while energy futures prices are generally unbiased predictors of future spot prices, there are certain notable exceptions. For both base and precious metals, the results are much less favorable to unbiasedness hypothesis. For precious metals and copper and lead, we strongly reject the null that β=1 at all three horizons. For the these other base metals, while we cannot reject that β=1, due to large standard errors. Finally, both corn and soybean futures have β close to 1, while wheat has β<1. Excepting oil and base metals, futures tend to outperform a random walk specification in out of sample forecasts.
The regression we run is:
st - st-k = β 0 + β 1 (f t|t-k - st-k) + ε t
Where st is the log spot price at time t, ft|t-k is the log futures price at time t-k that matures at time t. The resulting β coefficients at the three month horizons are displayed in Figure 2.
Figure 2: β1 coefficients, estimated via OLS. *** denotes significantly different from unity at the 1% level, using HAC robust standard errors. Source: Author's calculations.

Despite the bias in futures, along a RMSE dimension, futures outperform a random walk for most commodities, except for base metals (the out of sample period is 03M01 to 08M07). That being said, the outperformance relative to a random walk is seldom statistically significant.

This is a topic that has long interested me.  Fama and French have an earlier piece that looks at this same issue, albeit framed somewhat differently.

In theory, the expected difference between the futures price and spot price is a risk premium.  The risk premium for precious metals and natural gas appears to be negative.  That is, locking in the price today (removing risk) typically means a higher price as well.  If you're making these goods then selling short is a no-brainer.  There don't seem to be any positive risk premiums.  That is, storing your goods to sell later [without selling short on the futures market] is a risk that pays no reward for any commodity.

So why would the risk premia be negative?  One possibility is that these commodities are key inputs to the aggregate economy and price variability comes mainly from supply shocks.  That was probably once true for oil but now it seems demand shocks are driving things, which reverses the risk premium (see work by Killian).  Alternatively, they can serve as a hedge against financial and economic meltdown. That means buying commodities (going long) is what the market does to reduce their overall exposure to risk.

My dissertation, which I never published, argued that these negative risk premiums help to explain why natural resource prices haven't trended up over time.  A negative adjustment for risk means Hotelling's interest rate is probably close to zero.

I'm a little less confident than I used to be about this story.  But I still think there is some truth to it.  I really need to dust off that paper...


  1. Not sure I understand your dissertation thesis - I thought that for Hotelling's theory the relevant interest rate (/risk premium) was that available in broader financial markets? I.e. even if a resource extractor can't get a positive risk premium from holding natural resources, they could by buying a bond...

  2. Yeah, I could have been way clearer in this post. Sorry, this was really more of a note to myself than a genuine effort to communicate an idea.

    I'll try to give the intuition briefly.

    Take oil, for example, say back 10-30 years ago. Prices generally bounced around due to news about supply or potential supply in the future. The really big price spikes came from oil embargoes, wars, etc. If prices go up, those who own the oil are very happy. But the rest of economy is not so happy. A reduction in oil supply is a real shift inward in the productive capacity of the aggregate economy.

    Now asset pricing theory tells us that in equilibrium we should expect the risk premium to depend on the covariance with the aggregate economy. So, if oil prices go up when the aggregate economy goes down, going long on oil is like buying insurance for a bad economy. Insurance, like buying and holding oil, has a negative risk premium--it's an investment that loses money on average but you're willing to do it because it pays off most when you need it most.

    Assets like stocks are the opposite, they pay most when the economy is booming and you need the money least.

    Now Hotelling is just asset pricing for natural resources. It says prices should generally go up at the rate of interest. The thing is, for some basic commodities, the risk-adjusted rate is about zero. So prices don't go up in the long run.

    Today demand shifts are driving oil prices, and that changes things. But I think the story still holds for precious metals and natural gas where supply uncertainty remains significant.

    The same idea applies to climate change discounting. Investment in curbing global warming reduces the chance of a bad thing happening. It is an investment that pays off a lot of climate change is a really bad thing and an investment that pays nothing if climate change really isn't so bad. This means the risk premium is negative.

    Somehow many of my colleagues (especially environmental economists) seem to forget or overlook this essential fact. Weitzman gets it. So do the macro/finance guys who care about climate change. So does John Quiggin. Rank and file environmental economists don't get it, and it really shows in the literature.

  3. Ah, I understand now - thanks for the longer explanation.

    So what are the practical implications of the negative risk premium for investing in climate change prevention? (e.g., what mistakes would policy made based on the work of the "rank and file" be prone to?)

    And does the asset pricing analogy really hold for climate change or other environmental public goods, where the "proceeds" of investment are shared broadly, rather than captured by the investor alone?

  4. R:

    Yes, I think all of this matters a lot for climate policy. A big part of the debate in economic circles pertains to the discount rates that should be used in weighing the benefits and costs of regulation.

    If someone naively things the risk premium is positive, then the future--where all the benefits of climate change mitigation reside--gets discounted heavily. If we see it is actually a negative risk premium, then we discount the future much less.

    Strangely, depending on one's assumptions, all of which may seem reasonable, one can get discount rates anywhere from slightly negative to 10% or more. And these different assumptions lead to cost/benefit calculations for curbing C02 that range from HUGE, like approaching 100% of GDP to basically nothing.

    This ambiguity is not inspiring for economists...

  5. Nice - as an advocate of something being done about climate change sooner rather than later, I like this argument for why the financial discount rate should be low (for those who don't believe a lower "social" discount rate is appropriate).

    Back to the original point, the more I think about it, the more I struggle to believe that natural resource commodities (even aside from oil) have a negative beta. Natural gas, iron ore, copper, phosphate fertilizer, most ag crops - pretty much you name it, it spiked in mid-2008 and fell dramatically when the world economy nose-dived. I'm sure someone has actually run the numbers before, but it seems anecdotally that demand is a big driver for all of them and they're only occasionally countercyclical.


Post a Comment

Popular posts from this blog

Nonlinear Temperature Effects Indicate Severe Damages to U.S. Crop Yields Under Climate Change

Commodity Prices and the Fed