UPDATE (AND CORRECTION): Since I flavored this article with all things Chicago, let’s pretend that the University of Chicago hosted its conference at, well, the University of Chicago, not in New York City as I noticed on a belated reread of my sources.
The title refers to a consensus-shattering paper that was unveiled last month before a Who’s Who of economists and central bankers at a conference hosted by the University of Chicago.
Paul Krugman gave the keynote, but the meeting’s focus was on the paper’s authors—two Wall Street big shots, Morgan Stanley’s David Greenlaw and Bank of America Merrill Lynch’s Ethan Harris, and two academics, James Hamilton and Kenneth West. To keep it simple, I’ll call them GHHW.
The paper more or less shredded former Fed chief Ben Bernanke’s favorite defense of his quantitative easing (QE) programs—that QE lowered Treasury yields.
In fact, if you believe in the accuracy of the type of analysis GHHW conducted, QE may have actually increased Treasury yields. By parsing data and financial news more thoroughly than in prior studies, the authors found that yields rose, on average, when bond traders were presented with news about QE. (I recommend Hamilton’s blog write-up for a quick summary, although if you’re also looking for key charts, see Exhibits 4.11 and 4.12 on page 82 of the paper.)
But despite having the data to fully reverse the findings of other researchers, GHHW didn’t take it quite that far. (They were too polite for that.) Up against a strongly pro-QE crowd, they settled on the less ambitious conclusion that “the Fed’s balance sheet is a less reliable and effective tool than as perceived by many.” Between the lines, though, they painted a picture of QE being about as powerful as the host city’s passing game. (To save you the trouble of looking it up, Daaa Bears ranked last in the NFL.)
As far as pre-GHHW “perceptions,” the authors described a consensus that QE lowered 10-year Treasury yields by about 100 basis points, an amount they then refuted. That 100 basis point consensus is consistent with a few different literature reviews, as pointed out by GHHW, and also with claims by FOMC members. It went undisputed by the conference attendees who published their comments. (Three Fed regional bank presidents, an ECB Executive Board member and a few others delivered formal responses.)
The new research is important, in my opinion, not so much for academic reasons but because I think it foretells the future. Before I explain why, though, I need to insert a disclaimer about the likely accuracy of any study that attributes yield changes to QE news of one type or another.
That is, methods for establishing how much QE moved the bond market are essentially guesswork, even after GHHW’s improvements. Bond prices respond to traders and investors not only establishing new positions but also unwinding or rebuilding prior positions in combinations unknowable and for reasons derived from all past fundamental and technical information and ultimately also unknowable. Trades may occur because prices have gone up in the past, because they’ve gone down in the past, because the market is overbought or oversold, because a different market has become more or less attractive, because traders seek opportunities to lock in profits or cut losses, and for countless other reasons. As such, it’s easy to jump to the wrong conclusion by attaching a single fundamental cause to every price change—there’s no such thing as a sequence of single-cause price changes, and even if there were, we could only guess at the causes.
Why GHHW Upsets the Playing Field
All that said, let’s acknowledge that some researchers’ guesswork is better than others. I suspect GHHW are closest to the truth, partly because they were more careful than others, but also because a different type of result predicts their conclusions. I described that result in “QE’s Untold Story,” where I showed that commercial banks and broker-dealers extended credit between QEs by just as much as the Fed extended credit during QEs, and that the two sources of credit growth alternated depending on whether QE was “on” or “off.” Here’s the key chart from that analysis:
As I described last year, the Fed grabbed the credit-growth baton for QE laps and returned it to commercial banks and broker-dealers for QE pauses, and whoever didn’t have the baton stood still, creating the “argyle effect” shown in the chart.
Unlike GHHW, “QE’s Untold Story” didn’t separate short and long rates (the data didn’t allow for that), but it challenges the orthodox narrative from a different direction. Namely, it says if you draw a circle around banks, broker-dealers and the Fed, the amount of credit supplied to everyone outside the circle appeared to be unaffected by QE. Whereas the orthodox narrative holds that those outside the circle were forced to chase a restricted credit supply, the data tell a different story.
Or, another way to say the same thing is that banks accepted central bank reserves as an adequate replacement for assets transferred to the Fed—they didn’t seek to replace those assets on a like-for-like basis—and that decision would have diluted QE’s effects on yields. Some banks may have even welcomed the chance to replace long-term assets that were mismatched to their liabilities with different assets (central bank reserves) that carried no such mismatch risks.
Also, the federal government’s decision to lengthen its debt profile would have diluted potential QE effects as well, as noted by GHHW and others.
So plenty of other evidence shows why QE didn’t work as planned (it paints the bigger picture behind GHHW’s findings), it’s just that economists haven’t paid much attention to it. Macroeconomists, in particular, are known for reaching hasty and unrealistic conclusions, so it’s not surprising that they might paper over the holes in their QE studies or rely on theories that ignore the true mechanics of bank credit, which makes it difficult to grasp the relevance of the data in “QE’s Untold Story.” Getting banks right is especially important (see my articles “Learning from the 1980s” and “An Inflation Indicator to Watch” or for a fuller discussion, my book Economics for Independent Thinkers.)
As Hamilton wrote on his blog, “Our study raises a caution about the event study methodology. There is a potential tendency to select dates after the fact that confirm the researcher’s prior beliefs about what the effect was supposed to have been.”
In other words, economists tend to fit the “facts” to their theories rather than the other way around.
Who Might GHHW Have Been Thinking Of?
Hamilton didn’t name names, but consider that Ben Bernanke spent much of QEs 1, 2 and 3 selling the very conclusions GHHW debunked. Have a look at these excerpts from Bernanke’s speeches during his last few years at the Fed:
- “Securities purchases by the central bank affect the economy primarily by lowering interest rates on securities of longer maturities.” (11/19/2010-1)
- “The evidence suggests that such purchases significantly lowered longer-term interest rates in both the United States and the United Kingdom.” (11/19/2010-2)
- “Purchases of longer-term securities have not affected very short-term interest rates, which remain close to zero, but instead put downward pressure directly on longer-term interest rates.” (2/3/2011)
- “Generally, . . . research finds that the Federal Reserve’s large-scale purchases have significantly lowered long-term Treasury yields. . . . Three studies considering the cumulative influence of all the Federal Reserve’s asset purchases, including those made under the MEP, found total effects between 80 and 120 basis points on the 10-year Treasury yield. These effects are economically meaningful.” (8/31/2012)
- “A growing body of research supports the view that LSAPs are effective at bringing down term premiums and thus reducing longer-term rates.” (3/1/2013)
- “The preponderance of studies show that asset purchases push down longer-term interest rates and boost asset prices.” (1/3/2014)
To his credit, Bernanke was crystal clear in explaining what he was trying to achieve and why he believes it worked. That made him an easy mark when GHHW, whether they intended to or not, took direct aim at his published positions. Bernanke needed their meticulous analysis like the North Side Gang needed Al Capone.
And with that background in mind, let’s look to the future.
What to Expect in the Next Deleveraging
In the next severe economic downturn (whenever it occurs), central bankers are likely to embrace QE as readily as Bernanke did. They’ll first lower the fed funds rate as much as they can, but then they’ll feel the pressure to do more. (Sidenote: GHHW also disparaged negative interest rates.) They won’t say, “Look, the economy has too much debt and the best thing we can do is be patient and, well, do nothing more than we’ve already done.” That would violate the principles of today’s hyperactive interventionism—the chattering classes accept few excuses for policy inaction, and not knowing if a policy does more harm than good isn’t among them.
So the question is this: When tomorrow’s quantitative easers succumb to the pressure to act, how will they explain their actions? Actions require narratives, and with GHHW having toppled Bernanke’s narrative with Chicago-strength winds, policy makers will need a new one. So what will it be?
I would say one narrative is more likely than any other—that is, QE fell short of the objectives only because it wasn’t large enough, and to work properly it needs to be absolutely massive. Future Fed chiefs will argue that you can always bring yields under control if you just buy enough bonds, and to some degree they’re likely to be right. Their new motto will be “the bigger the better.”
Maybe so, but it’s also exactly what New York Fed President Bill Dudley told us to expect in his response to GHHW. He said, “If LSAPs are not as powerful as some of the event studies imply, the answer is not to simply discard the tool, but instead to look for ways to enhance its efficacy and use it more aggressively (emphasis mine).” Dudley then touted open-ended asset purchases, commonly known as QE infinity.
And that’s not all. Consider the charts and speech by ECB Executive Board Member Benoît Cœuré, also delivered in response to GHHW. Cœuré showed that the Fed’s QE left about half of total Treasury issuance in private hands after accounting for foreign central bank holdings, whereas the ECB has soaked up so many bonds that private investors are left with possibly less than 10% of all German Bunds. He then shared data suggesting that the ECB bossed Bund yields by more than the Fed bossed Treasury yields, as you might have expected. He argued that the key to QE success is to use the oldest trick in market manipulation—buy such overwhelming amounts that everyone else has to forage for a puny remaining supply. (Alright, he may not have called it market manipulation, but the rest is an accurate summary.) In other words, Cœuré’s GHHW response was to pen an ode to QE domination, which seems the natural endgame.
To be sure, the central banking gods may have written a bigger QE into our future long before the University of Chicago convened their emissaries, but now their plans are even more clear. Expect the Fed to follow the ECB and Bank of Japan in making sure the next time it expands its balance sheet, it’ll achieve total domination. You might imagine the Fed’s balance sheet blanketing the bond market in a thick, full-length coat (thick enough to withstand those Chicago winds), one that’ll make the current balance sheet look ragged and threadbare by comparison.
And you might also expect the irony to be lost on central bankers such as Dudley and Cœuré. Can the cautionary advice in a paper titled “A Skeptical View of the Impact of the Fed’s Balance Sheet” really lead to a more aggressive use of that balance sheet? In fact, I think it will.