QE & Stock Prices – A Review of Recent Data

One of the primary goals of Quantitative Easing is the portfolio rebalancing effect and subsequent wealth effect that supposedly occurs as the Fed reduces outstanding private sector bond holdings and forces investors to chase returns up the risk spectrum into other asset classes to replace lost potential real returns.  I won’t discuss the aspects of “money printing” and “debt monetization” here (which I think the mainstream gets mostly wrong), but rather, I’d like to focus on the pure data between periods when reserves are increasing and stock returns.

We’ve now had over 4 years of varying forms of QE so the data is quite a bit more constructive than at times in the past when most of us were just speculating about QE’s impacts.  At first glance, the chart below might lead one to conclude that there is a very strong correlation between between QE and equity market returns, but the data is not so conclusive upon closer inspection.  This has been due to the way the programs were implemented in staggered steps over the years with periods of buying tending to be rather short.  Contrary to popular opinion, during the majority of the time in the last 4 years reserve balances have been stagnant or even declining.  On a weekly basis, reserve balances have been rising just 47% of the time and have actually been declining 53% of the time.

Since QE began in late 2008 there’s a correlation between reserve balances and stock returns of just 0.65.  But that’s not a completely fair look at the data.  There have been four distinct periods when reserve balances were surging or declining/stagnant over the last 4+ years.  I’ve broken this down more clearly below.

Periods when reserve balances were increasing includes (on a weekly basis):

9/10/2008 – 1/7/2009

2/11/2009 – 5/20/2009

6/24/2009 – 2/24/2010

11/17/2010 – 7/13/2011

Periods when reserve balances were declining substantially include:

1/7/2009 – 2/11/2009 (reserve balances fell by 31%)

5/20/2009 – 6/24/2009 (reserve balances fell by 12.5%)

2/24/2010 – 11/17/2010 (reserve balances fell by 20%)

7/13/2011 – 9/26/2012 (reserve balances fell by 15%)

What’s interesting here is that the correlation between stock returns and decreases or increases in reserve balances is not quite as clear cut as most presume.  See below for returns during reserve balance increases:

9/10/2008 – 1/7/2009 (S&P 500 DECLINED by 25%)

2/11/2009 – 5/20/2009 (S&P 500 INCREASED by 6%)

6/24/2009 – 2/24/2010 (S&P 500 INCREASED by 22%).  

11/17/2010 – 7/13/2011 (S&P 500 INCREASED by 11%).  

That looks pretty much as we might expect.  But things get more interesting when we look at periods when reserve balances were declining:

1/7/2009 – 2/11/2009 (S&P 500 DECREASED by -10%)

5/20/2009 – 6/24/2009 (S&P 500 INCREASED by 1%)

2/24/2010 – 11/17/2010 (S&P 500 INCREASED by 8%)

7/13/2011 – 9/26/2012 (S&P 500 INCREASED by 9%)

In other words, regardless of whether reserve balances were rising or falling, stock prices were mostly rising.  Even in the face of substantial declines in reserve balances over the last 4+ years the S&P 500 has risen.  Of course, there are a lot of moving parts in this data, but the correlation between stocks and reserve balances is not terribly clear when we get granular with the data.

( Chart via Orcam Investment Research)

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:
TwitterLinkedIn

  • LVG

    Stock prices are driven primarily by expected future earnings. And earnings have been driven sky high by the 10%+ budget deficit we’ve been running. If you want to know why stocks are rising look at the government spending and not the Fed’s magic show.

  • Barak

    Cullen, i don’t believe that is the correct way to look at it. since the market is forward looking you need to separate between programs with a scheduled end to those without a scheduled one. if participants think a program will end it is likely they will acte it in advance. so in order to do this analysis correctly one needs to build a program that tries to find the timing that shows the best correlation between the data sets and observes what is the value for that correlation. since the market is reacting to many different inputs even a correlation of 0.7 is substantial for that matter.

  • SS

    I am not positive, but Cullen could be covertly responding to the many articles lately written about how the loan to deposit ratio is fueling speculation with new money. If Cullens findings are right then it casts serious doubt on the idea that investors are getting new funds through QE and putting it into the market.

  • Tom Brown

    Cullen,

    In your old article that you link to above:

    http://pragcap.com/the-fed-still-not-monetizing-the-debt

    I totally agree that QE in and of itself is an asset swap and is not Monetizing the debt.

    Bernanke, in the quote you provide says

    “In contrast, we are acquiring Treasury securities on the open market and only on a temporary basis, with the goal of supporting the economic recovery through lower interest rates. At the appropriate time, the Federal Reserve will gradually sell these securities or let them mature, as needed, to return its balance sheet to a more normal size.”

    Is it true that if the Fed holds a Treasury (rather than a private investor), then the government does NOT have to pay interest on it to the Fed?

    Secondly, what exactly happens when a Treasury is held to maturity at the Fed? Forgive me for my ignorance, but I’ve never purchased a Treasury directly… does the government normally pay the face value of the Treasury back (the principal amount) to the investor at the maturity date? Assuming that’s the case, does the government actually have to pay this to the Fed if the Fed holds a Treasury to maturity?

  • Geoff

    Tom,

    The Treasury does pay interest to the Fed, but then the Fed remits it back to the Treasury. It’s ridiculous.

    Yes, the Treasury pays back the face value of the bond at maturity to bondholders, including the Fed.

  • GLG34

    Interesting. You dont see much comment on the declines in reserves in the media.

  • jswede

    I just did some work on this — one conclusion I drew, in regards to QE1 vs QE2, is that the market *learned* from QE1 and applied it to QE2.

    Stocks did not see QE1 for what it was – a Fed (Bernanke) put – until months into the operation, then stocks strongly reversed and rallied right up to the end of QE1 – started dropping literally the week after it ended.

    For QE2, Stocks reacted prior to the operation, starting with the Jackson Hole Speech, a full 2 months before QE2 purchases began. More interesting, the rally began to fizzle in April, almost 3 months before the well-communicated end of QE2 operations.

    So it would only make sense that correlations would be off, as the market was learning something new — the power of the Bernanke Put…

  • jswede

    should add, this ‘put’ effect was also in place during Twist: though we did not see a Reserves / Fed balance sheet rise, we saw stocks rise due to said ‘put’, further impairing any correlation study

  • Tom Brown

    “Yes, the Treasury pays back the face value of the bond at maturity to bondholders, including the Fed.”

    Does the Fed then “remit” this back to the Treasury?

  • http://www.nowandfutures.com bart

    Bernanke himself admits that QE is basically money printing, as in the Fed itself credits PD reserves.

    http://www.thedailyshow.com/watch/tue-december-7-2010/the-big-bank-theory

  • Tom Brown

    That is a funny clip.

  • Geoff

    Tom, I think the Fed only remits “profits” back to the Treasury, such as the net interest on the T-bonds. The principal repayment just disappears as the Fed balance sheet shrinks like George Costanza in cold water.

  • Tom Brown

    …however, by itself, QE is an asset swap. Now of course since the Fed has an unlimited ability to expand their balance sheet (I think!) they could use that method to “buy the Earth” as the MMers sometimes say (but at HIGHLY inflated prices of coure). But then what would Ben do with an over-priced “Earth” on his balance sheet?

    I’m wondering though when the Fed buys gov debt (and it could, in principle, buy ALL of it) and hold it to maturity, does that constitute a backdoor means of gov self finance? Apparently the interest payments that Treasury makes on that debt gets remitted back to Treasury (according to Geoff above)… but the Treas does need to pay the principal back at Maturity. Now even if that principal doesn’t get remitted back to Treas, this process could continue forever. I guess Cullen would point out that the banks are still involved since they must purchase the Treasuries originally at auction. And the Fed isn’t the gov so I’m assuming some kind of coordination between the Fed and the gov that may not be fair. And of course, as in all kinds of deficit spending, inflation is the true limitation.

  • Tom Brown

    Yes that makes sense. Nice image!

  • Mark A. Sadowski

    In late 2010 David Glasner published a paper showing that TIPS spreads and stock prices became highly positively correlated around 2008.

    http://krugman.blogs.nytimes.com/2011/02/02/inflation-and-stock-prices/

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1749062

    Previously the correlation had been rather weak. The implication is that the stock market began to root for higher inflation at about the time when the US economy began to suffer from a serious AD deficiency. More that two years later this correlation persists:

    http://www.themoneyillusion.com/wp-content/uploads/2013/01/Screen-Shot-2013-01-26-at-10.29.05-AM.png

    It’s obvious to nearly everyone that there is a strong relationship between QE and inflation expectations:

    http://thefaintofheart.files.wordpress.com/2012/09/merrier_1.png

    So it seems to me that looking at reserves is precisely the wrong way to go about this.

  • http://www.nowandfutures.com bart

    “I’m wondering though when the Fed buys gov debt… does that constitute a backdoor means of gov self finance? ”

    In my opinion, unquestionably yes – it’s Accounting 101. The banks, as in mostly the Primary Dealers, are very much involved but only as a middleman between the Fed and the Treasury (and the public).

    As far as the Fed remitting back to the Treasury, that is true but they do keep about 6% of their profits per the original Federal Reserve Act of 1913.

  • William Morthel

    Forgive me in advance if I insult your quantative abilities, but the stated correlation between reserve balances and stock prices of 0.65 looks high to me. Did you use CHANGES in reserve balances and stock prices (rather than ending values) to come up with this figure?…one would expect any two upward trending series to exhibit significant correlation based on actual values (i.e. my age and stock prices). As such, the change in the data is used in an effort to eliminate such biases in the data. My guess is that the true correlation (based on changes in reserves and prices) is actually much lower.

  • William Morthel

    With the way I spelled “quantitative” in the previous comment, I may have also insulted your literary sensibilities!

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

    Hmmm, I might not be running it the way you prefer, but using flat QoQ (for instance) changes actually brings about a negative correlation….

  • Anon

    agreed SS – I think it remains a widespread (incorrect) belief that the reason why QE is hugely market-positive is because banks are being given new cash and then putting this cash to work by buying shares. If this were the case, you would expect there would be a very high correlation between QE and market direction.

    The data point I would point to is:
    2/24/2010 – 11/17/2010 (reserve balances fell by 20%)

    If QE = new money going into the market, then its remarkable that new money can be withdrawn by a staggering 20% and yet the market rise by 8%!!

  • Andrew P

    The Treasury pays the Fed the interest, and the Fed returns 94% of its profits back to the Treasury at the end of the year.

  • Andrea Malagoli

    Correlation is not the only measure of dependency. The two time series look reasonably co-integrated. It is hard to negate that there is a general trend between the two.

    Also, stocks responded to the european LTRO as well. I believe that if you include that you’ll see a closer relationship.

  • Andrea Malagoli

    … not to mention that rising asset prices is one of the Fed’s stated goals …

  • Andrea Malagoli

    … and one more (sorry) … all it takes is the mere ‘mention’ of QE to get stocks to rise. See e.g. Abe’s rhetoric and Dragi’s jawboning in the past year.

  • William Morthel

    At the risk of beating a dead horse, I am afraid it is more than a matter of ‘preference’–biases in the data must be addressed, otherwise the potential for bad conclusions swells.

    Anyway, intuitively, doesn’t a low or negative correlation seem more consistent with the observation that reserves have been declining 53% of the time and only advancing 47%?