Yesterday’s stock surge off the lows was attributed to comments by several Fed officials who said they are now in favor of the Fed implementing QE3 if the economy continues to deteriorate. I think it would be useful to review what QE2 did and what it did not do.
Now, before we begin it’s important to understand that markets are highly complex dynamical systems. No single policy is going to control these complex systems and it’s impossible to understand whether certain policies would have had differing impacts if implemented differently (or not at all). Therefore, we can only work with the facts we have and the reality that we see before us. This data will work within what has actually occurred and not within what may or may not have occurred without QE2 (such a study would be useless as its findings would be unsubstantiated).
Last year around this time I said quantitative easing would be a great “monetary non-event”. This was based on the idea that QE, as it was being implemented, lacked a transmission mechanism which would allow it to substantially impact the real economy. My prediction that it would matter very little to the real economy and my ideas that it was not “money printing” or “debt monetization” were all met with a great deal of controversy as these ideas were well outside of the mainstream beliefs. With a full year of data in the bag we can judge QE2 pretty definitively.
What QE2 did not do:
It did not help house prices:
It did not reduce the cost of a conventional 30 year mortgage:
QE does not appear to have substantially altered corporate bond rates:
It did not cause the consumer to go on a spending binge:
It did not cause an increase in hourly earnings:
It did not cause businesses to go on an investment binge:
It did not work through the traditional monetary policy channel of increasing loans:
It did not cause real growth to surge:
This is all consistent with my initial argument before QE2. My beliefs were based primarily on the idea that there was no real transmission mechanism through which QE2 could positively impact the real economy. It would not alter the net financial assets of the private sector, altering bank reserve balances would not increase lending and it would not work through portfolio channels in any positive manner due to the balance sheet recession. I believed the result would be little to no real impact, ie, being a “monetary non-event”. But before we make any sweeping conclusions let’s look at what QE appeared to do.
So what did QE2 do?
We know definitively, that the program was misinterpreated by most as “money printing“, “debt monetization” and other terms that implied that hyperinflation or even high inflation were on their way. It’s pretty clear now that that is not the case, but that didn’t stop markets from reacting strongly. We know, definitively, that speculative actions in the markets increased. (Some of these data points are controversial, but again, we are working within the reality that we can see and not what we believe to be true. Correlation does not equal causation, but this should provide us with a fuller picture nonetheless.)
NYSE Margin debt exploded at a 35%+ YOY rate during the program:
It also appears to have helped stabilize the equity market:
We all know that commodities including oil and gasoline prices surged during QE2. I don’t believe it’s a stretch to assume that the massive increase in leveraged speculation coincided with an eagerness to protect one’s self from what was believed to be a highly inflationary environment. This is consistent with a surge in commodity prices.
It helped fuel higher headline inflation:
Gasoline prices played a particularly important role in the surge in inflation:
QE also appears to have had a negative impact on the US Dollar. This is consistent with the idea of monetary easing:
What’s so interesting about all of this data is that the real world data appears to have deteriorated while the data from some markets appears to have improved. In commodity markets we saw massive price increases that did not coincide with subsequent growth in the US economy. Now, some of this effect could be due to overseas growth, but as the BOJ recently reported, I don’t think we can entirely rule out the idea that investors substantially increased their speculative bets during the QE2 program which influenced market prices. The margin data would appear to confirm such thinking.
What we got from QE2 was essentially one huge margin squeeze on the economy as investors protected themselves from inflation via their market hedges (helping contribute to cost push inflation via commodity prices), but saw little to no real-world impact (no offsetting substantive increase in hourly earnings). The result was an increase in inflation and inflation expectations, but no positive follow-through in the real economy to offset the negative effect of the cost push inflation.
Interestingly, this policy gets right to the heart of the discussion that was started the other day by Scott Sumner with regards to MMT and its intense focus on the real-world. Mr. Sumner said:
“I wasn’t able to fully grasp how MMTers (“modern monetary theorists”) think about monetary economics (despite a good-faith attempt), but a few things I read shed a bit of light on the subject. My theory is that they focus too much on the visible, the concrete, the accounting, the institutions, and not enough on the core of monetary economics, which I see as the ‘hot potato phenomenon.’”
This is the line of thought that leads so many to misinterpret QE and its effects. Efficient market thinkers make little distinction between the real economy and the markets. Markets after all, are believed to be good representations of the real economy. This feeds into beliefs like “wealth effects” and the idea that the Bernanke Put is good for the real economy. But as I’ve previously described, markets are anything but efficient. And the markets most certainly are not perfect reflections of the real economy.
I won’t rehash all of the actual monetary operations and the operational realities of QE, but the critical flaw in QE2 is that it had no real transmission mechanism through which it could fix a balance sheet recession and solve real world problems. Instead, it’s largely based on the myth of “wealth effects”, altering real interest rates (which most consumers and business have little to no awareness of), portfolio rebalancing effects and altering expectations.
Clearly, with stocks substantially higher than they were a year ago and the economy more fragile than it was when the program started, I think we can confirm that Robert Shiller was indeed correct about the effect of an equity market wealth effect – it’s highly overrated.
Altering real interest rates and portfolio rebalancing are fanciful sounding in an academic study, but in the real world consumers and business owners have little perception of real interest rates. Particularly during a balance sheet recession. What alters consumer spending habits is their spending desires relative to real earnings. In the case of QE2 we saw a decline in real earnings and a jump in inflation. This is consistent with consumers experiencing a reduction in their standard of living and it is not surprising that we have seen very weak consumer data in recent quarters as a result.
What primarily alters business investment is whether or not they have customers walking in their doors. Because QE did not alter the net financial assets of the private sector (it is merely an asset swap) it did not provide consumers with more spending power which would lead businesses to increase investment. More importantly though, real rates are most effective when they cause a releveraging effect in the economy. And herein lies one of the primary problems with monetary policy during a balance sheet recession. Consumers don’t want to take on more debt! So businesses might refinance, but without the increased business there is no reason to expect them to increase investment. These facts are abundantly clear from the above data on private investment and personal consumption.
Altering expectations appears like pie in the sky economics to me. I don’t deny that there is a certain level of truth to the idea that animal spirits play an important role in the economy, but they cannot be altered via the Fed bond purchases as we experienced during QE2. The problem here is that the majority of consumers and businesses make very few of their daily decisions based on the fact that the Fed might be buying some more bonds. Because this operation does not alter the net financial assets of the private sector there is very little reason to believe that it will filter through the economy in any sort of meaningful way. So, as I’ve often said, QE2 was implemented in a manner that is similar to telling your blind child that he/she might become a world class archer one day. It builds up hope, but doesn’t follow through with any real fundamental effect that will help the child achieve the dream you have implanted in his/her mind.
The various Fed banks and several academic studies have been released over the duration of the program that focus on the events themselves. These “event studies” include data on the days when the NY Fed was actually involved in buying bonds. I have previously argued that these studies are nothing more than datamining based on efficient market hypothesis. But markets are far more complex than this and are not nearly as efficient in their price discovery as some might think. An example of event studies might lead one to conclude that a particular stock’s earnings do not matter because the stock tends to decline in price on the day that they report their earnings. Of course, that would be a nonsensical thing to conclude, but a carefully devised event study could be framed in such a manner so as to provide credence to such a theory. These “event studies” ignore substantial market data over the course of QE2 and imply highly efficient markets. This is grounds for deep skepticism in my opinion.
In terms of its real effects QE2 could have actually been more of a drag on the economy than a form of stimulus. We know for a fact that the Federal Reserve turned over $73B in profits to the US Treasury in 2010 alone. That is largely interest income that is being taken away from the private sector as a result of their massive balance sheet expansion. Remember, this is interest income that the banks could have been earning. Instead, they are receiving 0.25% paper in exchange for their much higher yielding securities. QE does not add net financial assets to the private sector so the net financial drag appears to have been quite substantial.
It’s also important to quantify the effect of the surge in commodities when compared to the fiscal stimulus enacted at the end of Q4 2010. We now know that Ben Bernanke believed that QE2 was necessary to help finance the extra spending. Of course, MMTers know this is nonsense as an autonomous monetary issuer in a floating exchange rate system never “finances” its spending. So, what’s interesting is that QE2 could have actually resulted in a tax hike via commodity prices. Back in February I mentioned that there was a likelihood that gasoline prices would surge into the summer months. This was worrisome because gas prices had already rallied into this strong seasonal period. At the time I wrote:
“If gasoline prices were to average $3.75 by this summer it would be the equivalent of wiping out the entire tax cut that was recently passed. If prices were to surge back to their 2008 highs it would be the equivalent of a $182B tax on the consumer since QE2 began.”
This indeed happened and it the consumer is now feeling the pinch. It appears as though QE2 may have actually contributed to offsetting the entirety of the payroll tax cut enacted at the end of last year.*
In sum, it appears as though the positives (wealth effect, portfolio rebalancing and lower US dollar) were more than offset by the many negative trends that persisted. I am a bit surprised by the fact that some Fed officials are weighing another go at QE. The data appears undeniably weak arguing in favor of further “experimental policy”. We have had our experiment and it did not work. What is the point of trying it again? And have we considered the possibility that it could actually makes things worse? As a risk manager, this looks like an awfully bad bet to me. Granted, Dr. Bernanke isn’t in the business of portfolio management, but he is in the business of creating price stability and full employment. I don’t see how this program helps him achieve these goals.
What about QE3?
Now, the Fed could implement QE3 in various ways that would differentiate is from QE2. They could pin long rates and essentially define an inflation rate (this is essentially what the quasi monetarists are arguing in favor of, however, they reject the notion of the balance sheet recession so I think they’re still missing the key element in this economic downturn). Or they could buy other securities. I have trouble concluding that the risks here outweigh the rewards. There is no need at this juncture for the Fed to purchase other securities from the banks. Playing market maker in 2008 was effective. I said it would be at the time. But this is a very different environment. We don’t need the Fed to stabilize the mortgage market. What we need now is real help to the US consumer. Buying more securities at this juncture will only further increase the profits to the Fed which will reduce the net interest income to the private sector. This is entirely unnecessary at this juncture.
The other strategy that is often discussed is pinning long rates. This would certainly “work”. By “work”, I mean that the Fed can achieve any rate across the curve that it desires. All it needs to do is name a price and not a quantity at a specific duration and tell the market that it will protect this rate. The failure to do this has been one of the primary arguments I have used against QE2 since its inception. QE2 could never alter rates meaningfully because it was implemented incorrectly by targeting size and not price. But the risks with this approach are enormous in my opinion. Imagine the market’s response to the idea that the Fed is a willing buyer of however many bonds it needs to buy to achieve a 2% 10 year rate? If you thought the speculative ramp after QE2 was bad I hate to imagine what would happen after this. The debt monetization and money printing articles wouldn’t come off the presses fast enough. This could, in my opinion, only exacerbate the margin squeeze we have seen in recent quarters.
What can be done?
At this juncture, I think we have to recognize that monetary policy has failed us. This does not mean that monetary policy has been entirely ineffective. It just means that it has been far less effective than other possible tools. In terms of the various monetary tools, QE2 appears to have been a particularly ineffective policy response.
Dr. Bernanke would best serve the American people by going to Congress and explaining to them that we are suffering from an extraordinarily rare disease that he simply does not have the tools to combat. He should urge Congress to understand that it is impossible for the USA to “run out of money” and that this debt ceiling charade has been a failure to understand our monetary system. With very low core inflation he should explain to Congress that they can afford to help their constituents more. He should urge them to understand that we are nothing like Greece in that we can’t “run out of money”, but that austerity could make our economy appear similarly weak. He should urge them to pass an immediate full payroll tax holiday and help alleviate the burden on the debt laden consumer. It won’t solve all of our problems, but at this point it’s better than throwing more monetary policy at the wall hoping that it will stick. Perhaps most importantly, Dr. Bernanke needs to understand that further fiscal policy does not need the aid of monetary policy as QE does not serve as a “funding” source for the US government and could actually offset fiscal policy via other negative channels.
* This section was edited on 8/4/2011