Pardon the constant QE coverage, but this is by far the most important topic in the market these days so it’s important that we exhaustively research and hammer out the details on the program. During my research last night I came across this excellent paper out of Europe titled “Innocent Frauds Meet Goodhart’s Law in Monetary Policy“. It is one of the most thorough debunkings of QE I have come across. In this paper, they studied the QE program in the UK and came to important conclusions that are also applicable to the USA:
“An empirical analysis of the effect of reserves on lending was conducted; we do not find evidence that QE “worked” either by a direct effect on money spending or through an equity market effect. These findings are placed in a historical context in comparison with earlier money control experiments in the UK. Overall, we conclude that policies with serious flaws continue to be supported because of their link to powerful innocent frauds, which have great intuitive appeal and are widely upheld.”
This research confirms many of my personal findings thus far. QE did not boost lending, did not cause interest rates to decline and perhaps most importantly it did not have an equity market effect. From an operational perspective this can be confirmed because QE is merely an asset swap between the central bank and the primary dealers. Because it does not add net new financial assets, boost aggregate demand or influence interest rates substantially there is no reason for the program to influence the equity market. That is, unless, investors falsely believe the policy is expansionary or could drive equities higher. Eager buyers may drive prices higher, however, without a subsequent change in fundamentals there is no reason for the gains to last (because there was no fundamental reason for the gains in the first place).
I’ve attached a portion of the paper below, however, it’s certainly worth reading in its entirety. The evidence against such a policy is quickly mounting. In addition to the damning comments by Minneapolis Fed President Kocherlakota, the BOJ findings that QE does not boost aggregate demand or prices and my own work this is just one more clear case showing that QE is likely to be a non-event and is certainly no panacea for the economy (for more see here, here, here and here):
Thus we find no link between QE and its intended outcome. We may also probe the monetary mechanisms through which this outcome is to be achieved. The more sophisticated accounts of QE explain that it may affect spending via rising prices of corporate assets, which helps firms to collect money for investment in the markets. Benford et al. (2009: 3) write that companies who hold deposits obtained from the selling of gilts “may use the money to purchase other assets” and that “this process should bid up asset prices.” Congdon (2009) likewise writes that QE will cause “an excess supply of money and an excess demand for equities, which will put upward pressure on equity prices.” And “once the stock market starts to rise because of the process just described, companies find it easier to raise money by issuing new shares and bonds.”
Below we address the empirical question of whether this occurred; first we note a logical problem with this scenario. The equity market effect assumes “an excess supply of money and an excess demand for equities” (Congdon 2009) pushing up equity prices and investment, and so purchasing power and spending. But this reasoning seems to suffer from a fallacy of composition. It may be possible for one firm to raise money and spend it by investing, so increasing other people’s deposits and their purchasing power. But this does not mean that total purchasing power has thereby increased. For where would the money come from that companies would collect with new IPOs? Even if higher stock market prices led to IPOs, and the money raised would be used for spending (i.e., it would be purchasing power), the question is where does the money come from used to buy the new equities? There are three possibilities—the source of finance could be:
(i) money in circulation, so that net purchasing power is not affected;
(ii) new lending; or
(iii) money obtained by selling other equities (or securities generally)
In the case of (i), net purchasing power is not affected. For (ii) to work, we should see a link between QE and new lending (which we did not; see figure 2). That leaves (iii) as the only possible mechanism for QE to lead to an increase in purchasing power. However, this cannot exist in the QE account, since a general desire to sell equities is exactly the opposite of “an excess supply of money and an excess demand for equities.” There is no logical case for QE to boost spending via an equity market effect.
We now look into the empirical evidence on the equity market effect. The UK equity markets indeed rallied from spring 2009 onwards, and for the sake of argument we may put aside the logical problems and (partly) ascribe the rally in equity prices to QE. (One problem is then that equity markets returned to losses in 2010, supposedly undoing QE’s beneficial effects the moment it stopped.) But did QE cause more money to be raised by new IPOs being floated? The London Stock Exchange publishes IPO data in its New Admissions Summary (London Stock Exchange 2010). We summarize the figures in table 1, below, where we compare March 2009–February 2010 to the two preceding years, March 2007–February 2008 and March 2008–February 2009. For comparison, we also add March, April, and May 2010, where our observations series end.
We see that compared to previous years, in the QE year fewer IPOs were attempted, percentage-wise more IPOs failed to attract any money, among those that succeeded the average value of money raised was much lower, and more of it came from the three top IPOs. One would like to make a more complete analysis of determinants of IPOs, which is beyond the scope of the present paper, but such analysis is unlikely to overturn the conclusion from these data against the idea that QE caused more money to be raised in the equity markets. It may have caused a stock market rally, but no discernable knock-on effect on corporate issues, investment, and economic activity as in the QE storyline. This is understandable if corporate borrowing in the markets is driven by the investment climate, not by monetary policy.
Is it possible that the outcome was actually the opposite of the intended effect? We have argued above that QE may decrease rather than increase bank lending, and a similar argument may be made for IPOs, which is the public’s lending to firms. If anything, nonbanks are plausibly even more sensitive to the negative balance sheet effects of QE. A first perusal of the data in the last column appears to support this. In the three months after QE was discontinued (March, April, and May 2010, where our observations end), there were twice as many IPO’s per month than in the QE year, percentage-wise fewer failed, and, for those that succeeded, the average sum raised per IPO was larger. It should be noted that the three largest IPOs account for two-thirds of all money raised in these months; even subtracting these, money raised per month in March–May 2010 was £449.6mn, more than twice the rate that prevailed in March 2009 to February 2010 (£192.9mn, even including the top three IPOs in this year).
We have now traced the intended effects of QE—more bank lending and more equity issues—and found both without support in the data. In view of these findings, what is striking is that throughout the episode, no one seriously questioned the theoretical effectiveness of the mechanisms of QE as applied in the UK. Debate concerned only whether it would work in practice; that it could work in theory was not in doubt. Duncan (2009) reports that critics worried that gilts might primarily be purchased from foreign institutions, and thus that the money would be flowing abroad, not into the British economy (to which then Treasury Secretary Darling replied that this would still benefit the exchange rate). But no one questioned that increasing reserves was tantamount to increasing the money supply. Others urged the Bank to shift the focus of the scheme and buy more corporate debt in an attempt to stimulate corporate lending markets, but no one queried if the scheme would bring banks to increase lending and so create the additional purchasing power.
The facts are that bank lending—and therefore purchasing power and spending—did not in any perceptible way respond to gilt purchases under the QE program, and that it was much weaker in February 2010 than the BofE would have liked. The best one can say is that no one knows how bad things would have been without QE—or as Stephanie Flanders, the BBC’s economics editor wrote: “Quantitative easing may well have saved the economy from a credit-led depression. We will never know” (BBC 2010). Such are the ways in which innocent frauds are perpetuated. The last words on this—at least in this place—are therefore, perhaps fittingly, Galbraith’s:
The difficulty is that this highly plausible, wholly agreeable process exists only in well-established economic belief and not in real life. The belief depends on the seeminglypersuasive theory and on neither reality nor on practical experience. … How good this simple, painless design, free from politics in the hands of responsible and respectable professional public figures free from political taint. No disagreeable debate, no pointless controversy. Also, and uncelebrated, no economic effect. (Galbraith 2004: 41 and 43–44)