Doc Selgin's QE Eye Test

Quantitative Easing

QEEyeTestSome weeks ago, I made some critical observations concerning the Fed's contribution to  the recovery.  In particular, I complained that, despite the decidedly mixed and ambiguous results of empirical assessments thus far, the view that Quantitative Easing has been a smashing success seemed well on its way to becoming official dogma, if not a more generally-held article of faith.

Even so, I was taken aback by the off-hand manner in which Fed Vice Chairman Stanley Fischer declared a QE victory, in the course of a speech given two weeks ago at the International Monetary Conference in Toronto.   Did the Fed's policies work?  According to Fischer, "The econometric evidence says yes.  So does the evidence of one's eyes" (my emphasis).

In fact, as I've already noted, the econometric evidence concerning the effectiveness of QE is hardly decisive.  For one thing, most studies have looked only at the interest-rate effects of the Fed's purchases, without troubling to ask whether those effects translated into any definite changes in spending, output, and employment.  For another, the interest-rate effect estimates are themselves not to be trusted.   A fairly recent IMF study on "Foreign Investor Flows and Sovereign Bond Yields in Advanced Economies," for example, notes — en passant as it were — that, controlling for such flows, the Fed's large-scale asset purchases resulted, not in the 90-200 basis point decline in long-term rates reported in various other studies, but in a decline of just thirty basis points, which is peanuts.  Other studies may, in other words, have conflated the effects of the Fed's asset purchases with those of concurrent "flights" from lower-quality Eurozone securities to higher-quality Treasuries.

But why bother with fancy econometrics when one can simply refer, as Vice Chairman Fischer did, to the "evidence of one's eyes"?   Fischer, apparently, found in that evidence compelling proof that QE worked wonders.  Fischer actually mentions only one piece of evidence, to wit: the fact that "the recent inauguration of the ECB's QE policy seemed to have an immediate effect not only on European interest rates, but also on longer-term rates in the United States."  But here, as with other inferences drawn by looking at interest-rate movements, connecting the dots isn't nearly as easy as  Fischer supposes.

Evidence that some good has come from the ECB 's belated easing is, in any event, not evidence that the Fed's easing did any good.  Try as I might, I just can't seem to get my eyes to focus on any clear and unambiguous evidence that it did.  Has Fischer, I wonder, been looking at the same things I've looked at?  If so, is he perhaps looking through rose-colored glasses, or is it my own vision or prescription that's faulty?

With such questions in mind, I decided to put the matter to a test.  Call it Doc Selgin's QE Eye Test, or Eye QE Test, or whatever else you wish to call it.  The instructions are simple: eyeball the following charts, gathered from various internet sources, recording all sorts of basic information pertaining to Quantitative Easing on one hand and the post-2008 recovery on the other.  Then decide for yourself whether the evidence of your eyes agrees with Mr. Fischer's relatively sunny impression, or with my own much gloomier one.

Please don't misunderstand me: I am not saying that my QE Eye Test, or any eye test at all, is a good way to evaluate the effectiveness of the Fed's post-crisis policies.  On the contrary: I only wish to cast doubt upon Vice-Chairman Fischer's suggestion that one's eyes are all one needs to determine that those policies worked.  My own belief, FWIW, is that it's going to take a lot more fancy econometric footwork to arrive at convincing answers.  I just hope it doesn't take as long to come to a proper understanding of the Fed's role as it took following that other "Great" calamity.


Doc Selgin's QE Eye Test

1) Fed Assets and Bank Excess Reserve Holdings.

Source: Gold, Stocks, and Forex, November 12, 2014,

2) Growth in Monetary Base, M2, and NGDP.

Source: Ed Dolan, EconMonitor,

3) Nominal GDP Gap.

Source: Michael Robert's Blog,

4) QE and Treasury Yields.

Source: Calafia Beach Pundit, October 29, 2014,

5) Unemployment and Labor Force Participation Rates.

Source: Conversible Economist (Timothy Taylor), December 11, 2013,

6) Unemployment Using June 2009 Participation Rate.

Source: Sean Davis, The Federalist, January 10, 2014,

7) Employment as Percent of Population.

Source: Infinite Unknown, March 8, 2015,

8) Employment as Percent of Population, Comparison with Great Depression.

Great Depression Growth Employment to Total Population Ratio
Source: Rise Up, the System is Broken,

9) Real GDP: Actual and Pre-CrisisTrend.

Source: Cecchetti and Schoenholz, The Blog (Huffington Post),

10) Economic Output as Percent of Potential Output.

Source: Andrew Fieldhouse, The Blog (Huffington Post), June 26, 2014,

11) Comparison with Other Postwar Recoveries.

Source: Planet Money, March 7, 2013,


That's it.  If these pictures make you feel all warm and fuzzy about the great job the Fed has done, then so far as you're concerned, Vice Chairman Fischer's beliefs are vindicated.  If, on the other hand, you find yourself doubting that all those trillions of new dollars have accomplished anything, then you can either count yourself among the pessimists, or have Doc Selgin write you a prescription for some rose-colored lenses.

  • Milton_Hayek

    Well, who you gonna believe? Me or your own eyes?

    Duck Soup 1933

    Thanks for the charts, they indeed are a great eye test!

  • Juan Castaneda

    George, certainly I agree the charts above speak for themselves and more research will be needed to assess more rigorously the financial and real effects of QE in the UK, the US and the Eurozone. However, as you know it's difficult to address the counterfactual: what would have happened if central banks hadn't been involved in such unprecedented purchases of assets? Would broad money figures have fallen approx. by 30% as it did happen in the US from 1929 to 1932? Surely the effects of such a collapse in bank money would have had disastrous consequences in the real economy. Extremely difficult (if not impossible) to assess.

    On a related issue, the rather mild effect (if at all) of the huge monetary base expansion in the US and in many other well developed economies since the end of 2008 has proven the lack of a reliable relation between the monetary base and market liquidity broadly defined though the running of the conventional money multiplier. Remunerating commercial banks' deposits at the Fed didn't help for sure, but this was not the policy followed by the ECB and still broad money growth in the Eurozone has been rather sluggish to say the least. The textbook explanation of (bank) money creation has become useless perhaps because of the mess and enormous uncertainty caused in financial markets in the last few years, also in those years prior to the outbreak of the Global Financial Markets. I must say I was anticipating somehow inflation in view of the colosal increase in the ECB's balance sheet at some point and I was wrong; I didn't take sufficiently into account the 'uncertainty effect' reigning in financial markets at the time (of course there has been inflation in financial asset prices since 2009 but not major CPI inflation).

    Congratulations for the excellent posts published on this new Center's site!

    Juan C.

    • George Selgin

      "Would broad money figures have fallen approx. by 30% as it did happen in the US from 1929 to 1932?" Though the answer is of course uncertain, I very much doubt it. Despite all the loose talk about "runs" on "shadow banks," the episodes are hardly comparable. So called "shadow banks" aren't suppliers of money according to either the M1 or the M2 definitions of such. And there was no threat of any collapse to the "real" banking system. Some banks were in fact swimming in liquidity. Nor, in fact, did the Fed act to enhance system liquidity during the early, crucial phase of the crisis, when it was instead bailing out non- (including "shadow") banks. Later, when QE got going, it led to almost identical, offsetting changes in the base money multiplier. It seems likely that without QE the multiplier would have remained the same; perhaps it would even have increased somewhat. If the Fed was really interested in preserving the money supply and aggregate spending, it would probably have done better by leaving B unchanged while eliminating IOR.

      • Juan Castaneda

        Thanks George; we agree, the Fed's focus on the massive expansion of the base hasn't been very effective in terms of broad money growth and paying interest on banks' reserves didn't help at all.

        It's puzzling though, the ECB is being charging first a zero and then, since June 2014, a negative rate on banks' deposits (- 0.14%) and yet monetary aggregates (see M3 interannual rates of growth data in the chart below, from the ECB website) haven't really grown significantly until December 2014/January 2015, when it announced its QE programme.

        • Juan Castaneda

          (Sorry, my mistake: since June 2014 the ECB charges a -0.10% to banks' deposits, not -0.14%).


        • George Selgin

          I see no puzzle. For monetary stimulus to work, you need two things: (1) base expansion; (2) an interest rate on reserves that's sufficiently low to discourage banks from accumulating excess reserves. The U.S. has had (1) without (2). Until recently the Eurozone had (2) without (1)!

          I'm afraid I can't make much sense of your chart.

  • Jim

    His eye will be twitching when QE4 has to be pumped in to keep the bubble full of hot air.

  • Mike Sproul

    Except for "so far as your concerned", everything else is correct.

    • George Selgin

      Thanks, Mike. The boo-boo is all better now.

  • Pingback: Daily Money & Banking News Update: 06/15/2015 | Carl Menger Center for the Study of Money and Banking()

  • joebhed

    Doc G.
    Thanks. Great analysis.
    I guess "econometrics' are oddly like beauty – in the eye of the beholder.

    My view of the obvious failure of QE lies in the evolving mis-relationship between the Fed's Monetary base expansion without any concurrent similarity with M-2, which of course further portends the ever-present, if not expanding, GDP-gap.

    BUT <– and that's a big but, I don't fault the Fed for that. Rather I fault the money system.
    Being debt-based, it establishes this lack of a transmission mechanism in a balance-sheet recession, such that the public's debt-aversion cannot be overcome no matter how hard the CB pushes on its ineffective policy strings.

    Were the Fed, or the government, to be granted monetary policy 'resources' consistent with its mandate to close that well-presented GDP gap, then we could be seeing :effectiveness" become the major metric with which to judge 'public' monetary initiatives. (I know that's not your shtick, Doc.)

    In the first iteration of "Quantitative Monetary Easing", the Bank of Japan and the Japanese government first agreed on the success-metrics of the plan. IOW, (x) amount of QME would, theoretically bring about (y) changes to this, that and the other national economic statistic.

    Of course it never worked as planned for the same reasons as ours. But it was not an open-ended, "hope-for-the-best", shot in the dark that must eventually lead to these eye-of-the-beholder debates on its "effectiveness".

    The national economy cannot move forward without 'money', no matter your political suasion.

    What we need is a means to provide money without debt, which implies the end of fractional-reserve based lending, as called for by Wolf and Turner, a proposal which I fully support.


  • vikingvista

    Spooky. A string fell off my sleeve, and when I pushed on the end of it, it reproduced one of your charts exactly.

  • Pingback: Some Links()

  • daniilgorbatenko

    With respect to the labour participation rates, there is a very interesting contrast with the UK situation, where there has been no fall in labour participation rates. Probably it was because the total size of QE in the UK has been much less than in the US relative to the size of the economy.