This is a new all-time low for a Federal Reserve Bank.  The SF Fed is out with one of the worst cases of datamining I have ever seen in my life.  Their conclusion – not only is QE2 putting no upward pressure on commodity prices, but it’s actually causing commodity prices to FALL. They write:

“Our analysis does not provide evidence that Federal Reserve large-scale asset purchases fueled the rise in commodity prices. It shows that, despite the fall in long-term interest rates and the depreciation of the dollar, commodity prices fell on average on days of LSAP announcements. The effects were more pronounced during the first round of LSAPs. The results may have occurred because the LSAP announcements heightened investor concerns about risk or led them to revise downward their growth expectations. Thus, other factors, such as growth in emerging market economies, are more likely to be the main drivers behind the recent rise in commodity prices.”

I am not even sure what to conclude here.  This sort of analysis is not even worthy of refutation.  The Fed doesn’t help their cause when they allow nonsense like this to be published.  It is pure propaganda.  I can see how one can make a fundamental argument with regards to the fundamental transmission mechanism of QE2 and its relationship to fundamental price changes in commodities, however, to reject the powerful psychological impact of QE2 and the obvious correlation between prices and this policy is beyond absurd.  To conclude that it is putting downward pressure on prices is difficult to even take seriously.


Got a comment or question about this post? Feel free to use the Ask Cullen section, leave a comment in the forum or send me a message on Twitter.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:

  • Anonymous

    couldn’t agree more

  • LVG

    April fools, right? It’s not too late is it?

  • SS

    This is insulting to anyone with eyes.

  • Derfem

    Strange… Today is not April Fool’s Day….

    It’s more a denial of any responsability nor mistakes. Politicaly correct speaking.
    Bottom line: they know they are causing inflation in Commos Prices (it is FED’s goal), and they deny it in all ways.

  • Pete

    Did it mean that you have to create a bubble before it bursts? It eventually will burst, no?

  • John

    Soon the Fed will be beaming interleaved subliminal messages directly to our computer screens and tvs. Or do they already?

  • John

    Someone has apparently already taken control. How the hell did a link to an add for Cisco suddenly appear for the word ‘computer’ above?

  • John

    There goes again. This is scary.

  • Zebra

    SFBay is one of the sickest place on earth. skyrocketing house price, homo, and high crime rate in certain area, highest tax and gasoline price of the US.. and now with a dust of Japan radiation. they all do weird stuff to your brain.

  • Zebra

    still, thanks for the late April fool joke.

  • Greater Fool

    By that logic, why not increase large-scale asset purchases by ten fold? Heck, why stop there? Why not 100-fold? Then food prices will be a fraction of what they are now, we will once again pay $0.70 for a gallon of gas, and gold will once again cost $100/ounce.

  • tradeking13

    Yes, we need more and larger QE to bring down commodity prices further. :)

  • http://greshams-law.com Greshams-law

    Who knew that you could get paid for being a complete moron! All you have to do is get a job at the SF Fed! Seriously, can you imagine that there are grown up people with families, suits, ties and driving licenses sitting behind their desks all day in San Francisco doing this kind of ‘research’. Perhaps I’m being harsh, I guess most people’s minds would turn to mush if they don’t have to pursue a profit…

  • http://www.pragcap.com Cullen Roche

    You’re not being harsh. This is a really embarrassing piece of research.

  • Dmitro

    Take it easy guys. There in California weed got legalized recently, so what would you expect from those researches after a few good puffs.

  • Albatross

    They clearly want to take credit for rising stock markets, but don’t want to take credit for rising commodities prices. This is just shameless and further evidence that their asymmetric response to bubbles has led to institution wide cognitive dissonance.

  • percolator

    How am I not surprised by this “analysis”?

  • http://kiddynamitesworld.com Kid Dynamite

    this is beyond asinine. it’s as if the SF Fed just discovered what “sell the news” means!

    of course, their hypothesis is easily testable: the Fed should talk down the possibility of QE3 for the next several months, concluding with a June statement that QEII is over and that there will be no QE3.

    Then, the day after that meeting, announce an emergency meeting where they conclude that we need a trillion dollars of QE3 LSAPs. Anyone at the SF Fed want to take the other side of my wager that commodity prices will rise on that day’s surprise announcement?

  • George H

    Shouldn’t be surprised by the SF Fed. Their last research paper was QE created 3 million jobs.

  • Dylan Johnson

    Im a 19 year old business/economics major at Pitt. In october Andrew Sorkin from CNBC came to my university preaching non-sense that the government bailouts and stimulus spending was absolutely nessecary and more would only make things better i engaged in a argument with him during his lecture.

    I explained QE2 would allow speculators to run up commodity prices to new highs (based on pessimism of the Feds ability to get the U.S out of the recession) and this would obviously destroy consumer purchasing power, in turn nullifying the feds efforts to keep interest rates artificially low & economic stimulus spending to boost aggregate demand.

    I suggested that the money the consumer would have to spend from lower interest rates and tax refunds and stimulus would simply be spent on the same amount of commodity at a higher price, so in fact the Feds have done nothing to re-ignite growth in the economy. They simply have created no growth and now may induce another double dip recession or god knows what new financail/economic crisis by moreless creating a perfect environment for speculators to drive commoditys through the roof as they already have.


    ^^^ My article from january on the same subject, explains my view more in depth.

  • Dylan Johnson

    P.S sorry for grammer and mis-spelling, im in a class, trying to take notes and write this at the same time on my netbook. haha

  • Coolidge Low

    If this is not pure propaganda, then we had better revisit earlier PragCap posts on “Confidence”. It is all about confidence, lose it and it’s game over!

  • http://oldprof.typepad.com/a_dash_of_insight/ Jeff Miller

    Data mining might mean trying many different research approaches to find something that proves your point. It might mean leaving out part of a data series. It might me trying thousands of different variables and finding a handful that fit. I see no evidence of any of these, and it is usually pretty obvious.

    In fact, I am having trouble seeing what you do not like about the research design or the data. It is fine to be skeptical if the conclusion lacks face validity, but then you need to find a specific criticism.

    BTW, this is a research note by staff, not an official Fed pronouncement. Your statement that the Fed should not allow this to be published runs against everything we stand for in the blogosphere. Why shouldn’t we encourage and discuss intelligently research of all kinds, regardless of the conclusion?

  • Johnny

    Well, that was uncalled for

  • apj

    I suppose it’s worth repeating…..QE2 is a promise of low rates for an awfully long time. It has been enshrined in FOMC statements, but that is redundant. QE tells you that rising rates are a long way off, by definition. In a way, the current “promise” is similar in effect to the (unneccesarily elongated tightening) 2004 episode, just taking a different approach. When you promise low rates for an “extended period”, (and QE2 is a rubber stamp for this), or nod and wink it like 2004, you invite speculation simply because you are effectively guaranteeing stable funding levels for an “extended period”. It is not money printing, or money sloshing around finding its way into risk markets, it is purely the guarantee of impunity. I suppose we’ll have a whole lot of wingnut explanations of why commodities (and risk markets) are falling when QE2 ends, but the reality will be clear to those who have been thinking about the monetary system and the way markets work for “an extended period”.

  • JWG

    “[C]ommodity prices fell on average on days of LSAP announcements.” Did a freshman in high school write this for the SF Fed?

    “Correlation does not equal causation” is the first thing you learn in analytics.

  • Greater Fool

    But then you learn in Politics 101 that demgoguery trumps data every time.

  • http://oldprof.typepad.com/a_dash_of_insight/ Jeff Miller

    Cullen — I am disappointed with your response on several fronts.

    1) The idea that the Fed should quash research reports. It is insulting to your readers to accuse them of being “naive” when you are not able to discern the difference between official pronouncements and staff work. For a full response, see here: http://oldprof.typepad.com/a_dash_of_insight/2011/04/the-blogger-manifesto.html

    2) Data mining — So far you seem not to understand the technical meaning of this term. You sound like some one slinging it out there because it sounds good to say. How about an explanation, or else back off.

    3) I really don’t see your objection to the research design. They picked out some days to study. I’ll bet if they told you the design before you knew the result, you would have agreed that it made sense.

    As one of your SA followers, I strongly suggest that you give this one more thought. It strikes me as an impulsive post, aimed at a viewpoint that you knew everyone would embrace. That does not add value.

    Just a thought.


  • http://brianelwinpomeroy.posterous.com Brian Elwin Pomeroy

    Information, or rather, the timing of the information, is power. When the wealth of the world is all in a very few stacks, those of us in the know, will know how to fix it. Or, we could be proactive and join accounts and fight them with numbers. There are a few hundred of them and a few billion of us. We could take em.

  • FixedIncomeGuy

    Earth to Jeff! This isn’t some shoddy blog. This is the damn San Francisco Fed. Janet Yellen is two steps from being Fed Chief. Are you kidding me? You really think you and the other bloggers (no offense Cullen) are on the same level as the Fed banks? Get real. We hold our government accountable. They can’t just publish propaganda like this and get away with it. Freedom of speech might apply to the government, but they need to be held to a higher standard. They can’t just say whatever comes to their minds.

  • http://www.pragcap.com Cullen Roche

    Right. If i publish a bunch of BS no one cares. When our govt starts publishing BS it becomes a big deal.

  • vimothy

    Cullen: Are you objecting to event studies in general or this one in particular?

  • Bob T

    It appears to me that most people that have posted comments on Cullen’s post didn’t take time to read the entire Fed posting and just reacting to portion supplied by Cullen.

    If one reads the entire Fed article and reviews Fig 1 showing commodity prices vs industrial production and overlay it with periods of LSAPs, it certainly looks like the most dominate influence is industrial production and not the Federal Reserve large-scale asset purchases. However I do agree that saying “It shows that, despite the fall in long-term interest rates and the depreciation of the dollar, commodity prices fell on average on days of LSAP announcements.” does not add anything to the basic issue of cause for commodity price direction since these were short term events. However, Fig 1 after 2008 does show commodity prices rising at a slower overall rate vs industrial production than was occurring in 2007, whatever the cause(s).

    I personally have no clue if Fig 1 is accurate or not. Can anyone confirm or reject its validity? I sure do not want to come to an erroneous conclusion if data presented is incorrect.

    Bob T

  • vimothy


    But it’s an *event* study.

    If we wanted to understand the effect of earnings on stock prices, then we might estimate a linear regression model using a cross-sectional dataset. If we want to understand the impact of announcing earnings on stock prices, then we need to use an event study. In both cases we can control for factors other than earnings by including them in the model as explanatory variables. Your criticism seems misplaced.

  • http://www.pragcap.com Cullen Roche

    I don’t agree. Read their conclusion again:

    “Our analysis does not provide evidence that Federal Reserve large-scale asset purchases fueled the rise in commodity prices.”

    That should read:

    “Our analysis does not provide evidence that Federal Reserve LSAP fueled a rise in commodity prices during the specific days when the operations were being conducted”.

    Their conclusion is far too broad and vague and is a clear attempt to absolve the Fed based on a very narrow study. I see your point, but I think you’re reading into a bit too much.